You are on page 1of 77

Workplace stress measured by Job Stress Survey and

relationships to musculoskeletal complaints

Stefan Holmstrm

Department of Psychology
Ume University, Ume, Sweden
2008
Cover design by Camilla Hakelind and Linus Andersson
Copyright 2008 Stefan Holmstrm
ISBN: 978-91-7264-550-9
Printed by Arkitektkopia AB, Ume, Sweden
Abstract
Holmstrm, S. (2008). Workplace stress measured by Job Stress Survey
and relationships to musculoskeletal complaints. Doctoral dissertation
from the Department of Psychology, Ume University, SE-901 87
Ume, Sweden: ISBN: 978-91-7264-550-9

The main purpose of this thesis was to evaluate and test the Job Stress
Survey (JSS, Spielberger, 1991; Spielberger & Vagg, 1999), a self-report
instrument which assesses workplace stress. In the thesis a thorough
evaluation is made of JSS scales and items, and the relations to health,
particularly musculoskeletal complaints. The aim of Study I was to evaluate
the factor structure and the psychometric properties of a Swedish version of
the JSS. The instrument was distributed to medical service personal and
metal industry workers (n=1186). Factor analyses demonstrated a good
resemblance between the present version and the American original version.
The results also showed that the internal consistencies, as well as the test-
retest reliabilities of the scales are high, and the concurrent validity are good.
Study II examined work-related stress measured by JSS for the subgroups of
gender, industry workers and medical service personnel, and special attention
was given to the problem of differential item functioning (DIF) on these
subgroups. The main findings were that both gender and occupation has a
substantial impact on specific sources of work-related stress assessed by JSS
scales and individual items. The result of the DIF analyses showed no item
bias in the gender subgroup, but for the occupational subgroups there where
items showing DIF in two of the scales. These items do not jeopardize the
conclusions made on scale level since the number of items showing DIF are
too few to make an impact on the overall result on the different scales. In
Study III the relation between self-reported stress and health, particularly
musculoskeletal problems were examined longitudinally in two metal
industry factories. Results showed high levels of stress and musculoskeletal
complaints in these factories and significant and strong relationships between
the JSS scales and musculoskeletal, as well as psychosocial ratings. Lack of
Organizational Support was found to be more related to musculoskeletal pain
than Job Pressure. Longitudinal differences were found between the factories
and between different types of musculoskeletal complaints. The general
conclusions from the studies are that the present version of JSS shows a good
resemblance with the American original, and that JSS is a useful instrument
for studying relationships between stress and health.

Key words: Job stress, Job Stress Survey, work-related stress, musculoskeletal
complaints, metal industry, differential item functioning, medical service,
longitudinal.
Acknowledgements
It has been a challenge to complete this thesis, but also an interesting
experience, and this work was possible because of the assistance, patience,
and support for several individuals. I wish to express my gratitude to all
those who have helped me during the years completing this work.

First and foremost, I would like to thank my supervisor Bo Molander for


his contributions over the years, but also for his patience, support and
hospitality. I would also want to than my co-supervisor Margareta
Barnekow-Bergqvist, and the Centre for Musculoskeletal Research for
financing my studies. The collaboration with the trade union IF Metall
has been important for the realization of this thesis, and I would especially
like to thank Ulf Bjrk, Christer Eriksson, and Per-Erik Johansson for
your contributions.

I want to thank Marina Heiden, Bengt Lundqvist, Kenny Brnberg, Tale


Taksic, Maria Nordin, Greg Neely, Anders Kjellberg, Frida Krger-
Nygren, John Jansson, and Carl-Johan Olsson for providing helpful
contributions and comments on my work.

I also want to express my gratitudes to my friends and colleagues,


especially John Jansson, Carl-Johan Olsson, Camilla Hakelind, and Stefan
Sderfjell for your support and for all the fun time. Special thanks to John
who has supported me in my roll as a teacher, and for always having new
ideas On the Mind.

Finally, I thank my parents Roland and Anita for their continued support.

Villa til' I die!

Ume, April, 2008

Stefan Holmstrm
LIST OF PAPERS
This doctoral dissertation is based on the following articles:

I. Holmstrm S., Molander, B., Jansson, J., & Barnekow-Bergkvist,


M. (2008). Evaluation of a Swedish version of Job Stress Survey.
Scandinavian Journal of Psychology, 49, 277-286.

II. Holmstrm, S., Molander, B., & Barnekow-Bergkvist, M. (2007).


Gender, occupation, and item bias: Job stress assessed by Job Stress
Survey. Manuscript submitted for publication.

III. Holmstrm, S., Molander, B., & Barnekow-Bergkvist, M. (2008).


Work-related stress and musculoskeletal problems in metal industry:
A longitudinal study. Manuscript.
TABLE OF CONTENTS

INTRODUCTION............................................................................. 1
RESEARCH OBJECTIVES ................................................................ 3
THEORETICAL FRAMEWORK ...................................................... 4
Background to the concepts of stress and occupational stress ....... 4
Assessing occupational stress ..................................................... 7
Background and development of the Job Stress Survey ................. 9
Conceptual models of occupational stress..................................... 10
Person-Environment fit ........................................................... 11
Job Demand-Control-Support model...................................... 12
Effort-Reward Imbalance......................................................... 13
Transactional process model .................................................... 14
Summary of the occupational stress theories ............................ 17
Sources of occupational stress ....................................................... 18
Factors intrinsic ....................................................................... 19
Organizational roles................................................................. 19
Work relationships .................................................................. 20
Career development................................................................. 20
Organizational factors.............................................................. 20
Home-work interface............................................................... 20
Occupational stress and health ..................................................... 21
Cross-cultural perspective on occupational stress ......................... 25
Gender, stress and health.............................................................. 26
METHOD.......................................................................................... 28
Design........................................................................................... 28
Participants ................................................................................... 28
Data collection procedure ............................................................. 29
Measures ....................................................................................... 29
Job Stress Survey...................................................................... 29
Perceived Stress Questionnaire................................................. 30
Self-rated health....................................................................... 31
Statistics........................................................................................ 31
Exploratory factor analysis ....................................................... 33
Differential item functioning................................................... 33
SUMMARY OF STUDIES ................................................................ 35
Study I .......................................................................................... 35
Study II......................................................................................... 37
Study III ....................................................................................... 39
GENERAL DISCUSSION................................................................. 44
Severity and frequency ratings of stressors.................................... 47
Further evaluation and development of the JSS............................ 48
REFERENCES ................................................................................... 49
INTRODUCTION
Work-related stress and health, and especially musculoskeletal problems
are an increasingly common feature of modern life. Research has clearly
established a relationship in occupational stress between high levels of
stress and negative effects on health, productivity, and well-being of
employees (e.g., Hurrell, Nelson, & Simmons, 1998; Karasek & Theorell,
1990; Murphy, Hurrell, Sauter, & Keita, 1995; Quick, Quick, Nelson, &
Hurrell, 1997; Wright & Smye, 1996). Musculoskeletal problems arise in
virtually all workplaces, even though preventive measures, for example in
the ergonomic work environment has been accomplished. Risk factors for
developing musculoskeletal problems can be divided into three categories
of factors; (1) physical, (2) psychosocial, and (3) individual (Arins, van
Mechelen, Bongers, Bouter, & van der Wal, 2001). Reviews of studies of
the first two categories (e.g., Arins, van Mechelen, Bongers, Bouter, &
van der Wal, 2000; Arins et al., 2001; Hoogendoorn, van Poppel,
Bongers, Koes, & Bouter, 2000; Larsson, Sgaard, & Rosendal, 2007; van
der Windt et al., 2000) revealed that physical factors, such as repetitive
movements, awkward postures, high force demands, work posture,
vibration and psychosocial factors, such as low co-worker support, high
quantitative demands, low job control and low job satisfaction are of
importance. These factors might be associated with musculoskeletal
disorders directly, or as a consequence of raising the level of stress in the
work place. It is well known that high level of stress is related to
musculoskeletal complaints and pain (e.g., Lundberg et al., 1994; Siivola
et al., 2004).
Individuals and organizations face mounting problems of managing
work-related stress and illness, but they lack in understanding of the
nature of occupational stress (Williams & Cooper, 1998). There are some
problematic circumstances with the occupational stress field; one problem
is that the term stress is imprecise and misused. Another problem is the
lack of consistency in measurement tools. The measurement instrument
should also help raise awareness of occupational stress at both individual
and organizational level it should identify individuals who need remedial
help, and it should provide information for the design of appropriate
interventions (Williams & Cooper, 1998). The absence of valid, reliable
and standardized instruments make studies of occupational stress highly
problematic (Love & Beehr, 1981), and the lack of standardization makes
approaches to measurement across occupations and companies difficult
(Williams & Cooper, 1998). Creating measurement instruments which
can serve as comprehensive diagnostic tools and which can be used by all

1
employees within organizations and companies are a challenge and an
important task.
Job Stress Survey (JSS, Spielberger, 1991; Spielberger & Vagg, 1999)
is a measurement instrument which is constructed to measure generic
sources of occupational stress encountered by employees in a wide variety
of work settings, settings that often result in psychological stress (cf.
Jackson & Schuler, 1985). One of the objects of this thesis is to make a
thorough evaluation of the factor structure and psychometric properties
for the JSS, and a Swedish version of the questionnaire. Rick, Briner,
Daniels, Perryman, and Guppy (2001) have in there review of self-report
instruments found that there have been relatively little data published on
the psychometric properties of the JSS, and therefore it is hard to fully
assess the strength of the instrument. In studies the JSS instrument has
been shown to be related to factors such as job satisfaction, personality,
and health (Bongard & alAbsi, 2005; ORoark, 1995; Turnage &
Spielberger, 1991).
When instruments of the self-report type are designed or when they
are evaluated care must be taken so that measurements are appropriate for
different important subgroups (e.g., gender, age, occupation, occupational
level). This is important in order for no item to be more difficult,
discriminating, or easily guessed for one group than for another. If, for
example, there are items which differentiate in response patterns between
important subgroups it could be a sign of item bias known as differential
item functioning (Smith, 2002; Swaminathan & Rogers, 1990; Zumbo,
1999). In this thesis the participants in the different studies are employees
from metal industry and medical service sector. A further aim of this thesis
is to evaluate the relation to health, and, in particular, musculoskeletal
complaints.

2
RESEARCH OBJECTIVES
This research project includes studies with different aims as presented
below.

I. The main purpose of the first study is to evaluate a Swedish


version of Job Stress Survey (Spielberger & Vagg, 1999). The
focus is on the psychometric properties of the instrument and to
what extent it is possible to verify the original factor structure in
the present version.

II. In the second study the main purpose is to analyse a Swedish


version of JSS (Holmstrm, Molander, & Barnekow-Bergqvist,
2004; Holmstrm, Molander, Jansson, & Barnekow-Bergqvist,
2008) with special attention to the problem of differential item
functioning. Perceived severity, frequency, and overall
occupational stress for subgroups of gender and occupation
(assembly industry workers and medical service personnel) are
analysed.

III. The main purpose of Study III is to examine a Swedish version of


JSS (Holmstrm et al., 2004; Holmstrm et al., 2008) with
respect to relationships with health, in particular musculoskeletal
complaint over a four-year period.

3
THEORETICAL FRAMEWORK

Background to the concepts of stress and occupational


stress
One important thing in research is to possess knowledge of the history of
the actual fields of research, to understand how we got to the point where
we are now. It is important to emphasize that stress research developed
historically in two largely separate spheres, a physiological and a
psychological sphere (Mason, 1975a). Occupational stress as a field of
investigating job conditions and their consequences on performance and
health is a relatively new research field, which crystallized in the early
1970s. However, the roots of the occupational stress concept can be
traced back to Selyes (1936) research on animals, and Cannons (1929)
work on the physiological concomitants of emotion. In 1930`s Selye
discovered that a wide variety of noxious stimuli, such as exposure to
extreme temperatures, physical injury, and injection of toxic substances
evoked an identical pattern of physiological changes in the laboratory
animal he used. These noxious stimuli Selye later referred to as stressors.
Selye (1936; 1967; 1974) found that regardless of stressor the response
patterns were the same non-specific reaction. Some years later Selye
(1946) named this somatic response the General Adaptation Syndrome
(GAS), and defined stress as the non-specific body response to any stressor
and demand made upon it.
Cannon (1914; 1929) laid the base for the understanding of how
various emotional states affect physiological functions and diseases by
describing the fight or flight response. This response has played an
important role in stress discourse (Doublet, 2000). Another perhaps
equally important development by Cannon (1935), is the concept of
physiological homeostasis.
In the beginning Selye used the term nocuous agent but he felt that
it did not capture the real essence, and he looked for a more suitable term,
and Seyle adopted the term stress instead. Drawing on an engineering
analogy the non-specific response was the biological equal of what has
been called stress in lifeless objects. Mason (1975a) points to the fact
that it is quite difficult to trace Selyes own evolution about the use of the
term stress. At different periods during his development of the stress
concept, Selye has defined stress differently in terms of stimulus,
response or interaction between stimulus and response. From the
psychological paradigm came criticism. Lazarus (1977), for example,
argued that Selye had failed to take into account the psychological
signalling system, which distinguishes between dangerous events and

4
harmless ones. When the researchers in the field began to focus more and
more on the psychological factors of stress, Selye tried to generalise many
of his physical factors to psychological factors (Doublet, 2000) and saw
stress as a psychological response within the organism. Selye has admitted
that when he started working on the stress concept he gave little thought
to its psychological or sociological implication, because for him stress was
a purely physiological phenomenon (Selye, 1983). The physiological
approach to understand stress has had an enormous impact on stress
research. The work of Selye and Cannon has made important
contributions to the research field, and modern approaches to job-stress
assessment can be traced back to them and their work (Hurrell et al.,
1998).
Richard Lazarus has been one of the most important and influential
persons in the psychological development of the field of stress (Cooper &
Dewe, 2004). Lazarus and his colleagues noticed how one person who
reacted in a stressful situation did not necessarily want others to react in
the same way. To understand what happens between experiencing a
stressor and the reaction of the stressor one has to take variables such as
individual differences (e.g., goals and values), motivation, and earlier
experiences of similar situations into account (Lazarus, 1993). Lazarus
introduced the concept of cognitive appraisal, which is of importance for
understanding the differences in experience (Lazarus, 1966; Lazarus &
Folkman, 1984). This concept is central in the Transactional Process
Model, which will be reviewed in more detail later on. Cognitive appraisal
can be defined as the process which translates objective events into
perceptual experiences. This is an important contribution to the
understanding of subjective factors and how they can play a much larger
role in the experience of stress than objective factors.
Occupational stress or work stress has a relatively brief history in
scientific discourse. In work settings, stress can be produced by an array of
stressors, such as work tasks, psychosocial, and organizational stressors
(Murphy, 1996). The earliest definitions of work stressors were related to
role conflicts and role ambiguity. These stressors dominated the
occupational stress research in the 1960s and 1970s. The first self-report
questionnaire for measuring role conflict and role ambiguity was created
in the early 1970s. At that time the Person-Environment Fit model
(French, 1973; French & Caplan, 1972; French, Caplan, & Van
Harrison, 1982) was introduced and has become one of the most
influential and used models of stress (Eulberg, Weekley, & Bhagat, 1988).
The Person-Environment Fit model concentrates on the relation between
person and environment. The goodness of fit in these relationships are of

5
importance, and misfit in the relationship can lead to psychological stress
and ill-health (Caplan, 1987; French et al., 1982; Harrison, 1978, 1985).
There is another stress model which has had even greater impact in the
field of stress (Kristensen, 1995), namely Karaseks (1979) Demand-
Control model. This model is a synthesis of two research traditions; from
stress research the model inherited the notions of qualitative and
quantitative job demands, and from sociology it inherited the notion of
control (Karasek & Theorell, 1990). The intention with the Demand-
Control model was to provide a framework for analyses of the way in
which work is organised, and how it relates to alienation and ill-health
among employees. A further and more detailed description of these two
models will be presented later on.
The psychosocial contribution to the field of occupational stress has
helped creating a more complex picture of the field, through the dynamic
interaction between environmental and individual conditions. One
problem with studies of occupational stress is that they have been plagued
by confusing and inconsistent definitions of the variables of the stress
process. Seyle has been criticised for the label stress (e.g., Kasl, 1987a;
Mason, 1975a, 1975b), but he has tried to resolve differences and bring
clarity. Seyle stated once: I certainly did not discover stress, but only the
stress syndrome (1975, p. 38). Mason (1975a) regards it as remarkable
that the term stress has been so persistent, and given such a widespread
usage in biology and medicine in spite of disagreement over its definition.
Seyle (1976) agrees that the concept of stress is abstract, but even so he
thinks that the concept has an value. He makes a comparison with life:
Stress is an abstraction; but so is that of life, which could hardly be
rejected as irrelevant to the study of biology (Seyle, 1976, p. 49). Seyle
continues to say that there is no one who has studied life in a pure
uncontaminated form. It is always inseparably connected to something
else which is more concrete and seemingly more real (Seyle, 1976, p. 49) .
Kasl (1987a) is in agreement with Mason, and goes one step further,
thinking it is time to abandon the stress concept altogether.
There are limitations in current theories and measurement
instruments of occupational stress. One major problem is that most
theories of occupational stress reside in how occupational stress and strain
are defined and measured (e.g., Kasl, 1978; Schuler, 1980). If theoretical
concepts are ambiguous or undifferentiated, it is difficult to interpret
which aspect of job stress and strain that has been assessed (Vagg &
Spielberger, 1998). Different measures of stress assess different dimensions
of occupational stress, individual differences in ability, personality
characteristics, coping skills, and social support. These are all important

6
dimensions for understanding the impact of stressful job-related events on
individual and group level. Another issue in several measures of work-
related stress is that test items are lengthy or multidimensional in content.
As earlier remarked there are different approaches in defining stress.
This can lead to problems for researchers in different traditions. In fact,
scientists from medical and behavioural fields may have difficulty
understanding each other when discussing stress. This difficulty is mainly
caused by fundamental differences in the definitions of stress, even if both
traditions originate from the research of Seyle and Cannons. Despite the
confusions and inconsistencies between different traditions there are
important overlaps between them. Also, physiological stress and
psychological stress need completely different levels of analysis (Lazarus,
1966; Lazarus & Folkman, 1984). For example, what generates
physiological stress is not the same as what is stressful from a psychological
point of view (Lazarus, 1993). Zautra (2003) points to the fact that each
discipline has developed the concept using their own special meanings,
but has little knowledge of the complexities that exist in other traditions.

Assessing occupational stress


Occupational stress can be assessed in several ways, for example by ratings
made through observation, by measures of production, by self-reports, and
by interviews. The most common type of measurement when assessing
occupational stress is self-report measures (see e.g., Rick et al., 2001).
Examples of such instruments are Job Content Questionnaire (Karasek,
1985), Job Diagnostic Survey (Hackman & Oldham, 1975), Job Stress
Survey (Spielberger, 1991; Spielberger & Vagg, 1999), Occupational
Stress Indicator (Cooper, Sloan, & Williams, 1988), Occupational Stress
Inventory (Osipow & Spokane, 1987), Pressure Management Indicator
(Williams & Cooper, 1998) and QPSNordic (Dallner et al., 2000).
Questions have been lifted by some researchers (e.g., Frese & Zapf,
1988; Kasl, 1978, 1987a, 1987b, 1998) about the reliance of subjective
ratings when assessing occupational stress. Frese and Zapf (1988) propose
that there is a risk that trivial correlations occur in measures of self-
reported stress and health because of method variance or overlap between
the items in terms of content. These authors believe that the way out of
this problem is to use objective measures, and the use of objective
measures of stress would give a better and broader picture of stress in
working life. Critics on the other hand, state that measures of objective
occupational stressors are not often easy or even possible to obtain
(Hurrell et al., 1998), and Karasek and colleges point out that it would
take an outside observer much time to identify and evaluate the social

7
support situation of the worker (Karasek, Brisson, Kawakami, Houtman,
& Bongers, 1998). Several studies (e.g., Helmer, Barberger-Gateau,
Letenneur, & Dartigues, 1999; Lundberg, 2006; Miilunplao, Vuori, Oja,
Pasanen, & Urponen, 1997) have found that self-rated health measures
predict future health and life expectancy as well, or even better, than
medical examinations do. Kasl (1998) finds the discussion over subjective
and objective strategies of measures to be quite imprecise and
unnecessarily polarizing. The arguments from the discussion can be
summarized as follow, measures of self-report tend to be easily obtained,
cheaper, and more convenient. Whereas, objective measures, are more
expensive, clumsy, and difficult to obtain. The two measurement
strategies wont need to be competing with each other, they could give the
researcher important and complementing information, which help the
understanding and evaluation with an extended perspective on the
occupational stress in the actual organizations.
Measuring instruments in occupational stress can be categorised into
three groups, there are measures which assess job stressors, strains, and
health outcomes (Hurrell et al., 1998). Job stressors relates to a large
number of work-related environmental conditions regarded as having an
impact on the health and well-being of the employee. Strain is about the
employees psychological and physiological reactions to such exposures.
Health outcomes refer to negative health conditions thought to result
from exposure to job stressors.
It is not the demand or the source of stressor itself that is the issue; it
is the perception of that stressor (Cox, 1978; Lazarus & Folkman, 1984)
and the intensity of stressor (DeFrank, 1988; Dewe, 1989). Stressors
perceived as very severe and appearing with high frequency are more likely
to produce greater strain, thus contributing to the development and
progression of stress-related illnesses (Vagg & Spielberger, 1999). Failure
to take into account how often a stressor is experienced by an individual
increases the risk for an overrating of highly stressful events that occur
infrequently, while underestimating the impact of moderately stressful
events that frequently occurs. For example, if a co-worker is killed in a
workplace accident this would be extremely stressful for other employees
at the workplace. However, this highly stressful event is rarely experienced,
and therefore contributes relatively little to the overall experience of stress
at the workplace. In contrast, even though excessive paperwork is only
moderately stressful for most employees, it occurs with very high
frequency; therefore it contributes substantially to the overall experience of
stress.

8
Background and development of the Job Stress Survey
The Job Stress Survey (JSS, Spielberger, 1991; Spielberger & Vagg, 1999)
was designed to measure generic sources of occupational stress
encountered by employees in a wide variety of work settings. Each of the
JSS items illustrates a generic and job-related event, and JSS follows
Murphy and Hurrells (1987) plea for occupational stress instruments to
measure a core set of questions.
Prior to the constructing of the JSS, Spielberg and colleagues (Grier,
1982; Spielberger, Grier, & Pate, 1980; Spielberger, Westberry, Grier, &
Greenfield, 1981) had developed the Police Stress Survey (PSS), and the
Teacher Stress Survey (TSS) to measure work-related stressors
encountered by law enforcement officers and high school teachers. When
developing the PSS, focused groups consistent of police officers helped to
selected items from a large pool of items which had been derived from a
comprehensive review of police stress literature (Spielberger et al., 1981).
The PSS were used as a guide when the TSS was constructed, and from
the 60 items included in the PSS, 39 items were considered to be equally
applicable for work in schools and police work (Grier, 1982). The TSS
items were identical to corresponding PSS items, except that school and
teacher were substituted for department and police. The TSS items
were also reviewed and approved by an advisory committee of experienced
high school teachers. This committee also reviewed a pool of 50
supplementary items generated from the literature on teacher stress. From
the additional pool, 21 items were selected and added to the 39 items
adapted from the PSS to form the 60-item TSS.
The JSS instrument has been developed from the PSS and the TSS to
measure occupational stress (Spielberger & Vagg, 1999). There were 39
items that were found to be equally suitable for assessing job stress in both
teachers and police officers. Thirty of these items, which were judged to be
most applicable to a broad range of work situations, were elected and
adapted to form the JSS. The JSS items focus on aspects of work
situations that are likely to be encountered by workers in widely different
occupational settings, and in settings which are likely to result in
psychological strain. The selection of the items was based on extensive
research on the perceived severity and frequency of occurrence of stressors
experienced by teachers and police officers, but also by managerial,
professional, clerical, and maintenance workers (Grier, 1982; Spielberger
et al., 1980; Spielberger & Vagg, 1999).
This work fulfilled the Murphy and Hurrell (1987) request for an
instrument to measure generic stressors that facilitates comparing levels of
stress between different occupational groups, and the Jackson and Schuler

9
(1985) request that items should focus on aspects of work that often result
in psychological strain. Spielberg and colleagues also followed the
recommendation by DeFrank (1988) and Dewe (1989) that more
attention should be given to both severity and frequency of job stress.
Ratings of the perceived severity of a particular stressor give information
about the impact of the stressor event on the workers emotional state at
that moment. Examining the frequency of occurrence of a particular
occupational stressor provides traitlike data on how often the individual
has responded to that stressor. Spielberger (1983; 1988) views the
difference between the perceived severity of stressors events and how often
they are encountered as similar to differentiating between emotional states
and personality traits.

Conceptual models of occupational stress


The relationship between characteristics at work and employee well-being
has attracted considerable attention in the job stress literature. During the
years a number of conceptual models have been developed that relate job
characteristics to the health and well-being of working. It is among these
models we will find the theoretical framework for the JSS instrument
(Spielberger, 1991; Spielberger & Vagg, 1999).
The Person-Environment fit theory approach to stress (French, 1973;
French & Caplan, 1972; French et al., 1982) is one of the pioneer models
and also most widely accepted among organizational stress researchers
(Eulberg et al., 1988). Two other theoretical frameworks have been
particularly successful in generating and guiding job stress research: the
Job Demand-Control Model (Karasek, 1979, 1998; Karasek & Theorell,
1990) and the Model of Effort-Reward Imbalance at work (Siegrist, 1996,
1998). The Transactional Process Model (Lazarus, 1966, 1991; Lazarus &
Folkman, 1984, 1987) is a cognitive-motivational-relational theory, which
focus on the individuals own appraisals of different events or
environments (stressors).

Person-Environment fit
The Person-Environment fit (P-E fit) approach characterizes stress as a
lack of correspondence between characteristics of the person (e.g., abilities,
values) and the environment (e.g., demands, supplies). The central
hypothesis of P-E fit is that misfit between person and environment leads
to psychological, physiological and behavioural outcomes which can
increase the risk for morbidity and mortality (Caplan, 1987; French et al.,
1982; Harrison, 1978, 1985). Occupational stress research guided by the
P-E fit model has investigated a wide range of content, as, for example,

10
job satisfaction, job and organizational characteristics, individual
differences in attitudes, health status, employee skills and traits (Beehr &
Newman, 1978; Cooper & Cartwright, 1994; Jackson & Schuler, 1985;
Sharit & Salvendy, 1982).
The outcome of the P-E fit can take two separate versions. One
version focuses on the correspondence between personal skills, abilities,
and environmental demand (i.e., Demands-Abilities fit). The other
version focuses on the correspondence between environmental supplies
and personal values, motives, and goals (i.e., Supplies-Values fit). These
two types of fit (S-V and D-A fit) are often discussed together under the
heading of P-E fit (Dawis & Lofquist, 1984; French et al., 1982). The S-
V and D-A fits are fundamentally different, both in terms of their
underlying processes and their associated outcomes. The differences in
process are implicit in the components that constitute the S-V and D-A
fits (Edwards & Cooper, 1990). That is, the D-V fit suggests a process
where the individuals have to assemble their abilities and skills to meet the
demands of the environment. The S-V fit suggests another type of process
where individuals draw from their personal value structure to cognitively
evaluate the surrounding environment. These processes may be causally
related, as when an individual needs to satisfy environmental demands in
order to achieve valued states, but the processes are conceptually different.
There are differences in outcome. When environmental supplies
deviate from individual values, this can lead to dissatisfaction (Locke,
1969, 1976), negative affect (Diener, 1984), and other negative well-being
symptoms. In contrast, when environmental demands exceed individual
abilities, the individuals performance is likely to decrease (Hackman &
Oldham, 1980; Naylor, Pritchard, & Ilgen, 1980). In some cases well-
being may be directly influenced by the D-A fit, as when meeting
environmental demands is inherently valued by the individual (French et
al., 1982). The S-V and D-A fits are different versions of P-E fit, both in
terms of underlying processes and associated outcomes. Studies of P-E fit
often neglect these distinctions, and in some cases overlook them entirely.
This leads to theoretical and methodological research problems, which
limits the conclusiveness of the findings. The theoretical and
methodological problems comprise inadequate distinction between
different versions of fit, confusion of different functional forms of fit, poor
measurement of fit components, and inappropriate analysis of the effects
of fit (Chemers, Hays, Rhodewalt, & Wysocki, 1985; Edwards & Cooper,
1990; Fletcher & Jones, 1993).

11
Job Demand-Control-Support model
Karasek (1979) originally developed the Job Demand-Control (JDC)
model from conflicting findings in two research traditions. The JDC
model is a situation-centred model, and the model identifies psychological
job demands and job control (or decision latitude) as two primary job
sources which can lead to stress at work. Karasek defined job demands as
psychological stressors presented in the work environment (e.g., high
working pace, high time pressure, difficult and mentally exacting work).
Job control refers to the individuals ability to control his or her work
activities (Karasek & Theorell, 1990). Decision latitude consists of two
aspects, namely the width of skills used by the workers on the job (skill
discretion), and the workers authority to make own decisions on the job
(decision authority). Decision latitude is a term which has at times been
used as an alias for job control. Henceforth only the job control term will
be used which includes the decision latitude.
In the JDC model there are two different hypotheses, the strain
hypothesis and the learning hypothesis. The model assumes that having
control over the work process will reduce stress and increase learning,
while psychological demands will increase both stress and learning. The
strain hypothesis states that psychological strain and physical illness are
expected in the combination between high demands in work and low
control of the work process. The learning hypothesis states that high
demands in combination with high control leads to increased learning,
motivation, and development of skills. Another perspective of the JDC
model implies that control can buffer negative effects of high demands on
health and well-being.
In the eighties the model expanded with a social dimension (Johnson
& Hall, 1988; Johnson, Hall, & Theorell, 1989), and this expansion was
soon accepted by the creator of the JDC model (Karasek & Theorell,
1990). The Job Demand-Control-Support (JDCS) model acknowledges
that social support is vital for the employee when coping with different
demands at work. The social support dimension makes a distinction
between isolated jobs (jobs with few opportunities for social interaction),
and collective jobs (jobs in which interactions with co-workers are
integrated) (Johnson, 1989).
The evolution of the JDC model to the JDCS model has meant that
the two hypotheses have been slightly reshaped (Karasek & Theorell,
1990). The iso-strain hypothesis predicts that jobs which are
characterized by high demands, low control, and social isolation lead to
more illness, both psychologically and physically. The buffer hypothesis

12
states that job control and worksite social support moderate the negative
effects of high demands on the employees well-being.
Both the JDC and the JDCS models have been evaluated in many
studies. The models themselves, as well as the inconsistent empirical
findings of the models, have generated an impressive amount of both
theoretical and methodological criticisms (c.f., Kristensen, 1995). It has
been argued that the JDC(S) model is a male model because women are
less vulnerable to (iso)strain than men (e.g., Johnson & Hall, 1988).
These authors found that social support is a more important predictor for
cardiovascular diseases for women than work control. The argument
about male model is in line with Kristiansens (1995) comment that job
strain is another way of measuring social status. There are also a number
of different reviews of the models, and the results from them show that
the models are capable of predicting health as well as motivational and
productivity outcomes (e.g., see Bosma et al., 1997; de Jonge & Kompier,
1997; Hemingway & Marmot, 1998; Kristensen, 1995, 1996; Schnall,
Landisbergis, & Baker, 1994; van der Doef & Maes, 1998, 1999;
Verhoeven, Maes, Kraaij, & Joekes, 2003). A limitation of the above
reviews is that they are mainly based on cross-sectional designed studies.
For example, in van der Doef and Maes (1999) review, there were 53 of
the 63 studies cross-sectional designs. A cross-sectional design is ill-suited
to display the causal relationship between variables because it cant provide
evidence regarding the order of the variables. Therefore a longitudinal
design is needed in order to give evidence on the causal order of variables.
In a review based on longitudinal studies on the JDCS model de Lange,
Taris, Kompier, Houtman, and Bongers (2003) found support for
longitudinal causal main effects between psychosocial job conditions
included in the JDCS model and health and well-being outcomes (Cook
& Campbell, 1979; Taris & Kompier, 2003).

Effort-Reward Imbalance
Effort-Reward Imbalance (ERI) model (Siegrist, 1996, 1998) is an
alternative theoretical model, which looks at the reward rather than the
control structure of work (Marmot, Siegrist, Theorell, & Feeney, 1999).
The ERI model has put its focus on the centrality of paid employment in
working life. Effort at work is spent as part of a social contract exchange
process, where the society contributes in terms of occupational rewards.
Rewards can come in three different ways: money (e.g., adequate salary,
pay raise), esteem (e.g., respect and support), and security/career
opportunities (e.g., promotion prospects, job security and status). The
ERI model anticipates that an imbalance in cost and gain (i.e., a

13
condition of high effort and low reward) can cause stress, which can lead
to health problems (e.g., cardiovascular risks, sickness absence and poor
subjective health). For individuals, who have jobs which are demanding
and unstable at the same time, and who are not offered any promotion
possibilities, a state of stressful imbalance would be existing.
The ERI model makes an explicit distinction between personal
(intrinsic) and situational (extrinsic) components of effort-reward
imbalance. A combination of both the personal and the situational
components provide a more accurate estimate of experienced stress than a
restriction to one of these sources. The situational components consist of
efforts (e.g., psychological and physical demands at work) and the three
dimensions of occupational rewards (money, esteem, and security/career
opportunities). No specification is made of the health effect outcomes of
the different types of demands and rewards. It is rather the mismatch
between high cost and low gain received which matters most. The term
overcommitment is introduced in the personal components, a specific
pattern of coping with job demands and of eliciting rewards. This type of
coping defines a set of attitudes, behaviours, and emotions, reflecting
extreme striving in combination with strong desire of being approved and
esteemed. There is evidence that excessive efforts result from perceptual
distortion (e.g., underestimation of challenge), which in turn may be
triggered by an underlying motivation of experiencing esteem and
approval (Siegrist, 1996).
In empirical studies it has been found that high effort and low reward
at work is a risk factor for cardiovascular health, subjective health, and
mild psychiatric disorders (e.g., Bosma, Peter, Siegrist, & Marmot, 1998;
Peter, Alfredsson et al., 1998; Peter, Geiler, & Siegrist, 1998; Peter &
Siegrist, 1997; Tsutsumi & Kawakami, 2004; Van Vegchel, de Jonge,
Bosma, & Schaufeli, 2005).

Transactional process model


According to the Transactional process model (Lazarus, 1991), stress is a
result of an interaction between an individuals characteristics and
appraisals. These appraisals may consist of external or internal stressors
(e.g., events or environments), and the individuals internal or external
resources available to him or her at that moment. Stress is not a belonging
of the person, or of the environment. Stress is the result when there is
conjunction between person and environment, and only when this
transaction is evaluated by the person as harm, threat, or challenge to the
well-being of the person. Harm is about damage that has already occurred,
as in a loss of employment, failure when applying for work, or disapproval

14
by management or co-workers. Threat refers to a harm that has not so far
happened, but is expected in the future by the person. Stressors that are
apprised as challenging are conditions of high demands in which the
emphasis is on mastering the demands, overcoming obstacles, and
growing and expanding as an individual. In challenge, focus is on the
positive outcome and possibilities.
Appraisal is an essential concept in the Transactional process model,
and there are two central kinds of appraisals in the model, primary and
secondary. An individuals primary appraisal refer to the concerns of
whether or not there is any personal stake in the encounter (Lazarus,
1991), and it has been referred to as the motivational relevance part of an
encounter (Smith & Lazarus, 1990). Secondary appraisal refers to the
accessible options for dealing with harm, threat, or challenge. The
assignment of appraisal is to incorporate two sets of forces operating in
every adaptational transaction, namely, personal agendas (e.g., goals,
beliefs and experiences) brought to the transaction by the individual, and
the environmental realities that affect the outcome. Because personal
agendas vary between persons, and even within the person from moment
to moment, and since the environment is often quite ambiguous and
complex, persons attend selectively to what is happening, and evaluate it
in different ways. This can result in a great variation in the appraisals
individuals make in the same environmental context.
The coping process is a special form of appraisal. Lazarus and
Folkman (1984) define coping as the cognitive and behavioural efforts a
person makes to manage demands that tax or exceed the individuals
resources. Coping shall be seen as a process because the relationship with
the environment is constantly changing. Even if stable coping styles do
exist and are of importance, the coping process is highly contextual.
Effective coping styles have to change over time and across different
conditions (Folkman & Lazarus, 1985). The coping patterns also change
with age (Lazarus, 1993). Coping has a profound effect on psychological
stress and emotional states (Folkman & Lazarus, 1988). Coping is seen as
a mediator in the relationship between stress and illness (Carver et al.,
1993; Weaver et al., 2004; Wong, 1993). The Transactional process
model identifies different coping categories based on the intention,
function and meaning of coping efforts. For example, in the original
model there is problem-focused and emotion-focused coping (Lazarus &
Folkman, 1984), and there is meaning-based coping, introduced by
Folkman (1997).

15
Problem-focused coping refers to cognitive and behavioural efforts
used to manage or change the problem, and includes such strategies as
gathering information, planning, problem solving, and effort. If this
process is successful and alters the person-environment relationship for the
better, it will reduce or eliminate the psychological grounds for harm or
threat.
Emotion-focused coping refers to strategies, which only try to change
the way individuals attend to or interpret what is happening. For instance,
a threat that the individual successfully avoid thinking about, even if only
temporarily, doesnt bother the individual. Another strategy is reappraisal
of a threat into a no threat, meaning that the cognitive basis of the stress
reaction is removed. For example, if the employee can reinterpret a
negative comment from the supervisor as the unintended result of job
stress, the appraisal basis for reactive anger will dissolve. Emotion-focused
coping does not change the objective terms of the person-environment
relationship, but it changes how these terms are interpreted or attended to.
Denial, distancing and positive thinking can at times be useful techniques
for controlling psychological stress.
Meaning-based coping is expected to promote personal growth in the
midst of chronic stress. For instance, this strategy is the redefinition of
personal goals (e.g., stop drinking, in order to manage the job), which
may bring a new sense of purpose and direction into the individuals life. It
may also help the individual to sustain personal efforts for dealing more
efficiently with a stressor.
Coping is central to the stress process and its adaptational outcome. If
the individual makes the appraisal that this is an encounter he or she can
control or handle with his/her actions, problem-focused coping strategies
dominate. Emotion-focused coping predominates if the person judge that
nothing can be done to change the situation. Research has shown that
coping patterns vary between stressful encounters, and over time
(Folkman & Lazarus, 1985; Folkman, Lazarus, Dunkel-Schetter,
DeLongis, & Gruen, 1986; Folkman, Lazarus, Gruen, & DeLongis,
1986). Strategies, such as positive thinking are relatively stable across
encounters; this strategy is obviously influenced by personality.
Which strategies are good or bad depend on the situation and can vary
over time within a situation. For example, denial is widely seen as
unhelpful, but can be useful in certain circumstances. If a person
experiences symptoms of a heart attack, denial can be dangerous in the
sense that it can lead to delay in seeking treatment. The same coping
strategy can be successful during recovery from a operation after a heart

16
attack, but it can become unhelpful again if it continues for too long, so it
prevent the person to change lifestyle in order to prevent further attacks.
Research has shown that problem-focused coping is linked to an
increase of job satisfaction (Burke, 1998; Rick & Guppy, 1994) and a
reduction in anxiety and psychological stress (Grossi, 1999; Hobfoll,
Dunahoo, Ben-Porath, & Monnier, 1994). The results of the research on
emotion-focused coping give a contradictory picture, some studies show
that emotion-focused coping reduces somatic symptoms and psychological
stress (Parkes, 1990), and other studies show evidence of increased
psychological stress and psychosomatic symptoms (Day & Livingstone,
2001; Smri, Arason, Hafsteinsson, & Ingimarsson, 1997). Another point
discussed in research literature is the absence of clarity in the findings of
copings moderating role between different stressors and the outcome (see
e.g., Fortes-Ferreira, Peir, Gonzlez-Morales, & Martin, 2006). There
are some studies supporting this moderating effect (Day & Livingstone,
2001; Eriksen & Ursin, 1999). A second group have show weak support
for the moderating effect (Bhagat, Allie, & Ford, 1995; Greenglass, Burke,
& Ondrack, 1990), whereas a third group cannot establish any support
(Leiter, 1991; Rick & Guppy, 1994).
The Transactional Process model has been criticized by Brief and
George (1991) because it focuses too much on the individual employee.
They point out that it is just as important to identify stressful conditions
at work that have an effect on groups of employees. Other researchers
(Carver, Scheier, & Weintraub, 1989; Dewe, Cox, & Ferguson, 1993)
have argued that coping strategies should not be combined into broad
categories, but regarded separately.

Summary of the occupational stress theories


These conceptual stress models are the foundation for JSS from an
occupational perspective. The P-E fit model has been increasingly
criticized for lacking in specificity and failing to identify and quantify the
specific sources of stress put on the employees in the workplace (Chemers
et al., 1985; Edwards & Cooper, 1990; Fletcher & Jones, 1993). These
limitations have given rise to the development of models that
incorporated, expanded, and modified the concepts of P-E fit theory. Like
the P-E fit model, the JDCS and the ERI models focus on the interaction
of general demands of work, and the skills and characteristics of the
employees. This models give less attention to how specific job pressure
influences health and productivity. There are overlapping features in these
two models, and both models integrate psychological and sociological
theories to conceptualize and analyse relationships between psychosocial

17
factors and health outcomes (Rydstedt, Devereux, & Sverke, 2007). The
JDCS model primarily focuses on the work content and implies that job
task control is critical for understanding job strain, whereas the ERI model
is focused on the wider concept of reciprocity and fairness in the social
exchange process (Marmot et al., 1999; Siegrist, 1996). In the ERI model
subjective meaning of work experience is taken into account in the work
process through inclusion of individual characteristics (Calnan,
Wainwright, & Almond, 2000). The ERI and JDCS models contribute
with different but related types of information and aspects of the
relationship between the psychosocial work environment and health, and
the two models with their different features should be seen as
complementary to each other (Siegrist & Marmot, 2004). Rydstedt and
colleagues found that combining the two models will slightly increase the
explanatory power for predicting work-related mental strain (Rydstedt et
al., 2007).
There is a distinction between Karaseks JDCS model and Lazarus
Transactional Process Model in that JDCS asserts that the primary source
of work stress lies not within the individual, but in the characteristics of
the work environment. In the Transactional Process model stress is seen as
the result of the interaction between person and environment, and only if
the individual sees this transaction as a threat, the well-being of the person
is harmed or challenged. Lazarus conception of occupational stress and
theories concerning Person-Environment fit have both merits and
limitations, and can be seen as complimentary rather than contradictory in
providing a meaningful conceptual framework for understanding
occupational stress (Spielberger & Reheiser, 1994b). The major
differences between the two perspectives are that the Person-Environment
fit theories identify general conditions that produce strain at work (e.g.,
demands), whereas the Transactional Process model focuses on a
particular event and how the employee perceives and appraises this event.

Sources of occupational stress


To assess the consequences of stress in the workplace, we have to
understand the occupational sources of stress. Cooper (1986) has
differentiated six groups of primary work-related stressors: Factors
intrinsic, organizational roles, work relationships, career development,
organizational factors, and the home-work interface. These six categories
are not exhaustive descriptions of all potential stressors within each
category, but shall be seen as a useful framework for identifying physical
and psychosocial sources of job-related strain.

18
Factors intrinsic
Intrinsic work factors may have a serious negative impact on workers
physical health and psychological well-being (Cooper, 1987). There are a
number of factors that may be intrinsic to the job. Examples are poor
working conditions, such as lighting, noise, and smells (van Kempen et al.,
2002), work hours (De Raeve, Jansen, & Kant, 2007; Sparks, Cooper,
Fried, & Shirom, 1997), risk and danger, new technology (Cartwright &
Cooper, 1997; Korunka, Weiss, Huemer, & Karetta, 1995), work
overload (Warr, 1994; Westman & Eden, 1992), and work underload
(Melamed, Benavi, Luz, & Green, 1995).

Organizational roles
This category consists of three major aspects of organizational roles which
may be sources of stress: role ambiguity, role conflict and responsibility. A
wide range of events can create role ambiguity (Beehr, 1995). These events
can be summarized in three types of role ambiguity, lack of clarity about
scope and responsibilities of job, lack of clarity of objectives for role, and
inadequate information about work role. The stress indicators found to
relate to role ambiguity are low job satisfaction, physiological strain,
intentions to leave job, low self-confidence, low motivation to work, and
depression (e.g., O'Driscoll & Beehr, 1994). Role conflicts exist when an
employee is divided by conflicting job demands, or required to perform
tasks disliked, or outside of the job description. Role conflicts can be
manifested in different ways: (1) between an individuals internal
standards and required job behaviors; (2) between time, resources, or
capabilities of an individual; (3) because of conflicting expectations and
organizational demands; and (4) between multiple roles for the same
person which require incompatible behaviour (Quick & Quick, 1984).
Role conflicts can lead to decline in job satisfaction and increased anxiety
levels (e.g., Nystedt, Sjberg, & Hgglund, 1999). The connection
between role conflict and psychological stress is not as strong as that
between role ambiguity and psychological stress (Jackson & Schuler,
1985).
There are two types of responsibility in an organisation, responsibility
for people, and responsibility for things (e.g., equipment, buildings).
Responsibility for people has been found to be particularly stressful. Being
responsible for people often requires spending more time interacting with
others, more meetings and meeting deadlines. The interpersonal
relationships at work with superiors, peers, and subordinates can be a
major source of both stress and support (Makin, Cooper, & Cox, 1996).

19
Work relationships
Interpersonal relationships and lack of social support from others in the
workplace can be seen as a potential source of job-related stress. There is a
clear connection between negative interpersonal relations and the absence
of social support from within the workplace (Narayanan, Menon, &
Spector, 1999). Lack of support from supervisors is a major source of
stress at work (Cooper, Dewe, & Driscoll, 2001). Moyle (1998) has
shown that lack of support predicts lack of job satisfaction and mental
health. On the other hand, feelings of social support from colleagues and
supervisors may reduce psychological stress (Karasek & Theorell, 1990).
Leadership style is a potential source of stress at work for the employees.
For example, a leadership style which is task-oriented on the expense of
relationships and does not consider the needs, attitudes, and motivation of
the employees has been shown to create stress (O'Driscoll & Beehr,
1994).

Career development
Throughout the working life of the employee different factors act as
stressful. Examples are lack of job security, fear of redundancy, retirement,
and frustration of having reached ones career ceiling. Globalization of the
labour market, downsizing new technologies, reduction of levels of
management are factors which can be potential threats to the individual
employee and lead to feelings of job insecurity. Job insecurity may be one
of the most salient sources of strain for employees, and it will affect all
levels of the organization.

Organizational factors
Psychological strain attributed to organizational factors is often due to the
culture and management style adopted within an organization (Cooper &
Cartwright, 1994). Hierarchical, bureaucratic organizational structures
may lead to lack of employee participation in decision-making processes,
which can create a sense of no belonging and lack of job satisfaction in the
organization. Inadequate communication, especially between supervisor
and employees, may result in stress (Cooper et al., 2001).

Home-work interface
Managing the interface between work and various roles and
responsibilities outside of work is another potential source of stress
(O'Driscoll, 1996). Changes in family structures, increased participation
by women in the workforce, and changes through technological
development, for example, usage of mobile phones and portable

20
computers enable job tasks to be preformed in locations outside the office.
Together these changes have blurred the boundaries between work and life
outside of work for many workers, leading to a potential risk for a conflict
to emerge between home and work (Hill, Miller, Weiner, & Colihan,
1998).
Number of work hours is related to added risk of work-family
conflicts, decline in mental and physical health, and decreased family
functioning (Major, Klein, & Ehrhart, 2002). Mauno and Kinnunen
(1999) found that job pressure is negatively related to marital satisfaction.
Results from studies on work-family conflicts show that men had less
difficulty in combining work responsibilities and family relations than
women (Scott, 2001), and working fathers reported lower levels of work-
family conflict than working mothers (Hill, Martinson, Hawkins, &
Ferris, 2003). Parasuraman and Greenhaus (2002) stated that there are
relatively few studies which specifically have focused on work-family and
gender, and this represents a critical gap in work-family research.

Occupational stress and health


The effects of occupational stress on employees health and well-being
have been evaluated by several cross-sectional and longitudinal studies
which have shown that the combination of high demands and low control
at work is associated with psychological distress and health complaints
(e.g., Bourbonnais, Comeau, & Vezina, 1999; Cheng, Kawachi, Coakley,
Schwartz, & Colditz, 2000; de Jonge, Bosma, Peter, & Siegrist, 2000).
Other studies reports that low control at work is especially important (e.g.,
Elsass & Veiga, 1997; Stansfeld, Bosma, Hemingway, & Marmot, 1998).
Musculoskeletal disorders are another major problem in work and in
modern society, in spite of the ergonomic enhancements which have been
made at the workplace (Johansson, 1994). Musculoskeletal disorders in
neck, shoulder and back is the most usual type of occupational disease in
Sweden and in other countries (Kilbom et al., 1996). It is well known that
unfavourable psychosocial work conditions add to the risk of
musculoskeletal disorders (Arins et al., 2001; Bongers, Kremer, & ter
Laak, 2002; Bongers, Winter, & Kompier, 1993; Hoogendoorn et al.,
2000; Kjellberg & Wadman, 2007; Linton, 2001). Work-related
musculoskeletal disorders are not tied to a certain type of work. These
disorders affect workers in a wide variety of occupations, and they usually
take months or even years to develop (Hagberg et al., 1995). Because of
lack of standardized assessment criteria, estimations of the costs for work-
related musculoskeletal disorders are limited.

21
In a number of studies, social support has been found to be protective
against negative health outcomes, such as pain in neck and shoulder
(Arins et al., 2001; Bongers, Ijmker, van den Heuvel, & Blatter, 2006;
Larsson et al., 2007), back pain (e.g., Clays et al., 2007; Hartvigsen, Lings,
Leboeuf-Yde, & Bakketeig, 2004; Hoogendoorn et al., 2000), or
musculoskeletal problems in general (e.g., Woods, 2005). Social support
at work was one of the factors that predicted absentee from work among
patients with chronic musculoskeletal pain (Marhold, Linton, & Melin,
2002). For individual who already have problems with musculoskeletal
pain, access to social support was important in coping with symptoms and
related positively to health (Krause, Lynch et al., 1997). There are
different types of social support, and Grahn, Stigmar and Ekdahl (1999)
found in there study that social support at work was the most vital type of
support when helping with motivation for behavioural changes in patients
with long-standing musculoskeletal disorders.
Davis and Heaney (2000) state that the mechanisms for how social
support induces musculoskeletal ill-health remain undefined. One
explanation for this can be that most studies use computed scales when
reporting the results, presumably leading to the loss of more profound
information on the contribution of different components of social support
linking to health (Woods, 2005). Another explanation can be define in
definition of social support and how we group and categories social
support variables. There are some differences in how different researchers
group components of social support, and Woods (2005) uses six different
categories: (1) General social support, including support form supervisor,
co-workers and general acceptances by peers; (2) Good communication,
as, for example, ability to communicate with supervisor and co-workers;
(3) Satisfactory relationships at work, for example, satisfaction with
colleagues and work atmosphere; (4) Understanding of pain, that is,
supervisors and co-workers support of pain; (5) Help from supervisors and
co-workers when tings are difficult; (6) Social support away from work.
Cohen (2004) presents another categorization were he define social
support as intention to benefit an individuals ability to cope with stress,
and makes a distinction based on mechanisms between (1) Social support
(stress buffering); (2) Social integration (promotes positive states); (3)
Negative interactions (relationships as a source of stress). Cohen state that
these variables are associated with health outcomes, and that they
influence health through different mechanisms.
A wide range of occupational factors in the workplace can influence
the occurrence and severity of musculoskeletal disorders (Punnett & Gold,
2003; van Dien, Visser, & Hermans, 2003). The different factors can be

22
classified into two main categories, ergonomic (biophysical) and
psychosocial factors (Blair et al., 2003), and there is a variety of job factors
that affects the risk for developing musculoskeletal disorders. Ergonomic
factors are high intensity of muscular contractions, time constraints (e.g.,
high work rates, high frequency of mechanical loading), repetitive and
stereotyped work position or movements, static loading of long duration,
high precision movements of low intensity, and vibration. Psychosocial
factors, for example, are skill under-utilization, lack of control, lack of
support, unpleasant physical work conditions, and harassment or bullying.
Recent reviews have shown that there is a relationship between
psychosocial factors at work (e.g., low job control and lack of support,
high job demands and workload) and musculoskeletal symptoms (see e.g.,
Arins et al., 2001; Bongers et al., 1993; Hoogendoorn et al., 2000).
Nonetheless, the results are ambiguous and somewhat contradictory, and
the roles of the psychosocial factors in the development of the
musculoskeletal disorders are not yet clearly understood. Punnett and
Gold (2003) states that the relationship between musculoskeletal disorders
and psychosocial factors at work cannot be represented by a simple
exposure-response relationship, since these features at work often interact
with one another and also cause multiple types of musculoskeletal
disorders, even in the same work settings. They state that the
multifactorial nature of these relationships is sometimes misunderstood or
misinterpreted as a lack of scientific assurance that a cause and effect
relationship exists.
The ergonomic factors can act as stressors which can generate
musculoskeletal disorders through a number of pathophysiological
mechanisms (Blair et al., 2003), and the psychosocial factors have an
emotional value for the worker, which can influence the workers
subjective well-being in different ways (Hagberg et al., 1995). There are
different physiological explanatory models addressing the issue of how
work-related psychosocial factors can affect the development of
musculoskeletal disorders (Blair et al., 2003; Melin & Wigaeus Trnqvist,
2005). The Johansson and Sojka model (Johansson & Sojka, 1991) and
the Cinderella hypothesis (Hgg, 1991) are two of these models, in both
of this models the psychological stress effect on the mechanism are
recognized (Blair et al., 2003; Lundberg et al., 2002; Melin & Wigaeus
Trnqvist, 2005). But neither of the models is formulated in a way which
makes it easy to apply psychological self-rated data.
The Johansson and Sojka (1991) model relates to the finding that the
number of muscle spindles is high in the neck and shoulder region. The
muscle-spindle system is important for coordination of movements,

23
allocation of muscle activity, and for regulation of muscle stiffness. This
model assumes that long-time static or repetitive work causes dysfunctions
in the muscle-spindle system, which will increase the activity in the
system, which in turn increases the muscle stiffness and the production of
metabolites. Thus a positive feedback loop or vicious circle is generated,
where increased muscle stiffness may cause higher concentrations of
inflammatory substances (e.g., bradykinin, serotonin). This continues the
vicious circle by signalling back to the muscle spindles, which becomes
even more active. Increased activity influence the muscle spindles
regulation, which may cause dysfunctional coordination and inadequate
muscle activity. Such vicious circles can spread to other muscles than the
initial one, this distinguishes the Johansson and Sojka model from the
Cinderella hypothesis. The sympathetic nervous system (SNS) plays an
important role in the Johansson and Sojka model, as the SNS is involved
in vascular changes and changes in muscle contractility, in modulation of
afferent information from muscle spindles, and in metabolic changes at
cellular level (Passatore & Roatta, 2003). Thus, physical as well as
psychological stress is a main factor in the vicious circle process.
The Johansson and Sojka (1991) model has received support (e.g.,
Djupsjbacka, Johansson, Bergenheim, & Sjlander, 1995; Djupsjbacka,
Johansson, Bergenheim, & Wenngren, 1995; Pedersen, Sjlander,
Wenngren, & Johansson, 1997; Wenngren, Pedersen, Sjlander,
Bergenheim, & Johansson, 1998). One problem, though, is that most of
the support for this model comes from animal studies, and the model
needs more verification in human subjects (Knutson, 2000).
The Cinderella hypothesis is an analogy of the fairy tale with the same
name. Cinderella was the first to rise and the last to go to bed and was
thus likely to be fatigued in the long run. The Cinderella hypothesis
suggests that work-related myalgia is caused by overloading of low-
threshold motor units. During repeated or static load the small and
sensitive motor units are the first units activated in a muscle contraction
and they are also the ones which are the last to shut down when the
contraction is ended. These motor units are active as long as the muscle is
at work, and are only at rest during complete relaxation. These motor
units are therefore vulnerable to fatigue due to overload, which in the long
run can lead to metabolic disturbances and damaged muscle fibres.
According to the Cinderella hypothesis the muscle pain is due to
insufficient recovery, and findings in different studies support the
association between lack of muscle rest and myalgia in the trapezius (e.g.,
Hgg & strm, 1997; Sandsj, Melin, Rissn, Dohns, & Lundberg,
2000; Sjgaard, Lundberg, & Kadefors, 2000). Psychosocial factors are

24
suggested to keep motor units active, and Lundberg et al. (2002) have
found that some motor units which are active during physical activity are
also active during mentally encouraged muscle activity. In work which
demand low overall muscle activity, signals of fatigue are too weak too be
detected and the individual may continue working without knowing that
some motor units are overloaded. When the muscle exposed for static
load, the concentration of inflammatory substances and metabolites are
increased in the muscle, due to reduced blood flow, which results in an
increase in pain sensitivity (Lundberg et al., 2002). Psychologically
perceived stress is often more persistent than physical demands, which
indicates that psychological stressors are important contributors to fatigue
and overuse of motor units, even in the absence of physical demands, as,
for example, during breaks at work and outside of work.

Cross-cultural perspective on occupational stress


In a historical perspective, much of the research on occupational stress has
been performed in North America and Great Britain. More recently
important work is coming also from Northern and Western Europe
(Glazer & Beehr, 2005). However, it is unclear if the results and
conclusions drawn from studies in Western societies are generalizable to
non-western societies, such as China (Liu, Spector, & Shi, 2007). Cross-
cultural studies of psychosocial and health factors are important, because
such studies provide better understanding of variations of stress variables
and provide better opportunities for testing and evaluating theories and
instruments in different contexts (Marmot & Madge, 1987).
Results from cross-cultural studies show some evidence that similar
stress concepts may exist in different countries. Investigations of role
conflict, role ambiguity, and role overload in 21 countries showed that the
constructs and measures were valid with minor modifications (Peterson &
Smith, 1997). Liu and colleges (Liu et al., 2007) found that lack of job
control was a frequently reported stressor by employees in USA but not in
China, which were in line with their belief that Americans would expect
and experience more control. They found no difference between
university employees in China and USA in overall interpersonal conflicts,
but Americans reported more direct conflicts, whereas Chinese conflicts
were more indirect. The researchers concluded that this deviation came
from the cultural differences (individualism vs. collectivism). Spector et al.
(2004) found that managers in Anglo countries (Australia, Canada,
England, New Zealand and USA) reported more pressure in the work-
family interaction stressor, which was interpreted as a result of the longer

25
work hours for managers in Anglo countries than their counterparts in
China and Latin America.
Verhoeven et al. (2003) states that results from studies on job
demands, job control, and social support from colleagues and supervisors
may not be similar in different cultures and countries. A Polish study by
Makowska (1995) found no relationship between the JDCS model and
health. Along the same line a study of blue-collar workers in Japan showed
no relationship between the model and depressive symptoms (Kawakami,
Haratani, & Araki, 1992). Kushnir and Melam (1991) found in their
study on Israeli industrial workers support for the buffer hypothesis, but
no support for the strain hypothesis of the model in relation to somatic
complaints. The failure to confirm the JDCS model in these studies may
also be explained by other factors, but these results raise questions
concerning the validity of the JDCS model in different cultural contexts.
The JDCS model was created in north-western Europe and the United
States, and most of the studies using the JDCS model have been carried
out in that part of the world. One question which can be raised is thus
whether the model is capable of exploring relationships among work
conditions and health outcomes in countries and cultures outside north-
western Europe and United States (Verhoeven et al., 2003).

Gender, stress and health


In general women report more stress (Narayanan et al., 1999; Nolan-
Hoeksema, 1990; Spielberger & Reheiser, 1994a; Spielberger & Vagg,
1999; Vagg, Spielberger, & Wasala, 2002). Women reports pain more
frequently (Eriksen, Ihlebk, & Ursin, 1999; Eriksen, Svendsrd, Ursin,
& Ursin, 1998; Ihlebk, Eriksen, & Ursin, 2002). Musculoskeletal
complaints are more commonly reported by women in both general
populations (Hooftman, van der Beek, Bongers, & van Mechelen, 2005;
Leino-Arjas, Hnninen, & Puska, 1998; Linton, Hellsing, & Halldn,
1998) and in working populations (Gerr et al., 2002; Krause, Ragland,
Greiner, Syme, & Fisher, 1997). One explanation of these gender
differences could be less autonomy and fewer opportunities for
modifications of women jobs (Leino & Hanninen, 1995; Nordander et
al., 1999; Zetterberg et al., 1997). Another explanation is that women
react more powerfully than men to psychological stressors such as
interpersonal conflicts or deadline pressures at work (Treaster & Burr,
2004).
Psychosocial variables might be more contributing to musculoskeletal
disorders for women than for men, particularly to disorders in the neck
and shoulder region (Arins et al., 2001; Hooftman et al., 2005). One

26
predominant reason for this is that women work is more stressful for this
actual region than what men work is. Women work may generally be
perceived as being less demanding because it does not include forceful
exertion. It is more often characterized by a high static loading on the
neck and shoulder area. The work is often requiring fast paces, precision,
and recurrence for small muscles (Punnett & Herbert, 2000).
There are studies that report that men and women differ in the uses of
coping strategies. Men use more often problem-focused coping strategies
(Folkman & Lazarus, 1980; Hurst & Hurst, 1997), whereas women use
emotion-focused coping strategies (Hurst & Hurst, 1997; Tamres,
Janicki, & Helgeson, 2002). These differences in the use of coping
strategies disappear when occupation, education and position are
controlled for (Greenglass, 2002). This conclusion is supported by the
Long and Kahn (1993) findings that differences in use of coping strategies
come from results which are based on studies in which men and women
are not matched on occupation.

27
METHOD

Design
The studies in this thesis are exclusively based on self-ratings, and in the
questionnaires used, the subjective perspective of the participants was
asked for. In evaluation of JSS, cross-sectional designs were used (Study I
and II). In Study III the purpose was to examine the relationship between
work-related stress and musculoskeletal problems, and because a cross-
sectional design is ill-suited to display causal relationships, a longitudinal
design was applied.

Participants
Study I and Study II uses the same sample of participants for evaluation of
different aspects of JSS. Managers and representatives of labour unions in
metal industry and medical service units in northern parts of Sweden were
contacted and given the opportunity to participate in the work stress
project. Metal industry data came from eight different companies, varying
in size from around 50 employees to more than 1000 employees. In the
larger companies, only some of the departments of the companies
participated in the study. In the medical service sector, data were collected
from a university hospital and from health care centres, in several
communities. The university hospital data, as well as the data from health
care centres covered a representative selection of units and departments,
including units for administration and maintenance. A total number of
1186 employees participated, which gives a response rate of 79.0 % for
the sample. The age of the employees varied between 17.5 and 64.9 years
(M age = 43.1, SD = 11.1), the gender breakdown being 51.3 % men.
To evaluate the concurrent validity of the JSS instrument in Study I,
an additional sample of participants was used. These participants were 32
ambulance personnel, 29 men and three women, varying in age between
29.2 years and 58.1 years (M age = 40.6; SD = 8.31).
In Study III two of the metal industry factories participated in a four
wave longitudinal data collection, with the first in November 2001, the
second six month later, and then every 12 month. All workers from both
factories were asked to participate in the study. Table 1 presents employee
characteristics for the two factories over the four waves of data collection,
and for the employees who has participated in all four measurement waves
(T1-T4).

28
Table 1
Characteristics of Participating Employees in Factories A and B in each
Measurement Occasion, and of Employees Participating in all Four
Measurements (T1-T4)
F n tot M W Age range M age SD
A 54 73 51 3 19.1-63.6 37.3 11.89
T1
B 61 70 52 9 20.5-61.6 39.3 9.55
A 59 77 54 5 19.5-64.0 39.1 11.78
T2
B 57 73 49 8 20.9-58.1 37.8 9.83
A 53 70 48 5 20.8-59.9 38.5 10.51
T3
B 58 70 52 6 22.0-63.1 40.4 9.76
A 47 60 44 3 22.6-60.9 41.6 10.82
T4
B 56 70 47 9 19.7-60.2 40.9 10.03
T1- A 27 26 1 20.2-55.9 39.1 10.30
T4 B 29 24 5 23.9-56.1 38.4 9.35
Note. F=factory; n=Total number of participants; tot=Total number of employees; M =
men; W = women; M age = mean age (Study III, Holmstrm et al., 2008).

Data collection procedure


Data collections were conducted during working hours in the work place.
The participants were informed about the purpose of the study and
informed consent was obtained. Instructions were given about how to use
the scales of the survey, and there were several practice trials. If someone
wished to get help to understand the meaning of items, or wanted further
help about the rating procedure the test administrator provided such help.
Total time for instructions and completing the survey was approximately
one hour. After one to two months feedback was given to the work place
and results were presented on group level and discussed with managers
and employees.

Measures

Job Stress Survey


The JSS instrument consists of 30 items that express different job-related
stressor events (Spielberger, 1991; Spielberger & Vagg, 1999). The
instrument is divided in a severity part and in a frequency part. In the first
part the participants first rate the perceived severity of each stressor event
on a nine-point scale, ranging from 1 (low stress) to 9 (high stress). Every
stressor is to be compared with the stressor Assignment of disagreeable

29
duties, serving as a standard stressor and placed at the midpoint of the
scale (i.e., 5, moderate stress). For example, when the respondent is rating
a stressor like Poor or inadequate supervision, the respondent is
supposed to compare it against the standard stressor Assignment of
disagreeable duties. Severity ratings greater than, and less than 5 indicate
that Poor or inadequate supervision is considered by the respondent to
be more or less stressful than the standard. In previous research, this
standard has consistently been given an average rating by workers in a
variety of occupations (Grier, 1982; Spielberger et al., 1981). When
completing the first part, the respondent reports in the second part how
often (in days) each stressor occurred during the last 6 months, using a 10-
point scale from 0 to 9 or more.
The JSS instrument comprises three stress scales and six subscales. The
stress scales are named Job Stress Severity (JS-S), Job Stress Frequency (JS-
F), and Job Stress Index (JS-X). The JS-S scale shows the respondents
average rating of perceived severity of the 30 JSS stressor events. The JS-F
scale indicates the average frequency of occurrence of the 30 JSS stressor
events for the period of the last six months. The JS-X is the overall stress
scale for level of occupational stress. In the JS-X scale the severity and the
frequency ratings are combined into an overall indicator of perceived stress
(severity x frequency). By selecting only some of the 30 items, six subscales
can be created: Job Pressure (JP-F, JP-S, and JP-X) and Lack of
Organizational Support (LS-F, LS-F, and LS-X). Job Pressure consists of
the ten items which are attributed most directly to pressure of work, such
as meeting deadlines, working overtime and dealing with crisis
situations. Items reflect aspects of the jobs structure, design or duties.
Lack of Organizational Support involves ten items which reflect events
involving other people (co-workers and supervisors) or organizational
policies and procedures, rather than specific aspects of the job itself. Items
included are fellow workers not doing their job, inadequate support by
supervisor, and inadequate/poor quality equipment.

Perceived Stress Questionnaire (PSQ)


A Swedish version of PSQ (Bergdahl & Bergdahl, 2002) was used to
examine the concurrent validity of the JSS instrument. The PSQ measure
perceived stress by focusing on cognitive aspects more than emotional
states or specific life events (Levenstein et al., 1993). PSQ consists of 30
items, and items are scored from 1 to 4 (i.e., 1. Almost never; 2.
Sometimes; 3. Often; 4. Usually). The PSQ index is acquired through
computing the total raw score, using the formula (raw score-30)/90, and
varies from 0 (lowest level of perceived stress) to 1 (highest level of

30
perceived stress). Bergdahl and Bergdahl (2002) reported good concurrent
validity for the Swedish version of the instrument.

Self-rated health
The subjective health questionnaire consists of 28 items. When
responding to the health items a five-point scale, ranging from 4 (Yes,
absolute) to 0 (No, absolutely not) were used. The measure of subjective
health was done in two ways: First, participants assessed whether they
currently felt healthy. Second, on the remaining 27 items participants
were asked to estimate if they had experienced symptoms categorized as
being of psychiatric nature (e.g., worry, feeling of loneliness, difficulties in
sleeping), and if they had experienced symptoms of somatic nature (e.g.,
neck pain, stomach pain).
The questionnaire is the same subjective health questionnaire as used
in the Betula Study (Nilsson et al., 2004; Nilsson et al., 1997). The Betula
Study started in 1988, and is a longitudinal study with the aim of
exploring cognitive functions and health in adulthood. Four waves of data
collection have been completed in 5-year intervals since the start of the
project. The purpose for using the same questionnaire as in the Betula
Study is that the results from the Betula Study can be used as norms for
the questionnaire.

Statistics

Latent constructs such as stress give the researcher a delicate measurement


problem, as it is not always possible to observe the perceived stress of an
individual. Another problem is that there are many different definitions
and theories existing which describe the stress construct. There are several
theories that approximately describe the same psychological construct, for
example, theories directed at the interaction between person and
environment.
The empirical studies in this thesis are concerned with construct
validation of the JSS in one form or another. Factor analysis is one of the
most frequently used methods for evaluating construct validation
(Nunnally & Bernstein, 1994). Factor analysis describes a set of related
techniques rather than a single method, and factor analysis is a technique
with a long tradition in measurement theory. The goal of the factor
analysis is to find factors that explain covariation among different variables
and to illustrate how a small number of latent constructs (e.g., items that
measure stress) can explain covariation among larger numbers of measured
variables, such as behaviour ratings, scores on test items, or subtest scores

31
(Tinsley & Tinsley, 1987). Factor analysis can be divided into two classes,
exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).
Stevens (2002) clarifies the differences between EFA and CFA by stating
that EFA is used to identify the factor structure for a set of variables, and
to determine how many factors exist, as well as the pattern of the factor
loadings. CFA, on the other hand, is a theory-testing procedure, which
allows the researcher to specify the factor model in advance and define
which variables will load on which factors. In this thesis exploratory factor
analyses were used, the emphasis here is on replication, and on examining
if the same factor structure will unfold in the Swedish version as in the
original version of the JSS. The benefit with EFA is the possibility to
detect items with cross-loadings on more than one factor. Items with
cross-loadings on several factors are inaccurate and noisy and a threat to
the validity of the instrument (Rick et al., 2001).
Item bias is a major threat to the fairness and measurement validity of
the instrument (Kristjansson, Aylesworth, McDowell, & Zumbo, 2005).
Description of item functioning is a relatively modern psychometric
development (Nunnally & Bernstein, 1994), and analysis of item
functioning is of importance when evaluating construct validity (Cole,
1999; Dean, Holst, Kreiner, Schoenborn, & Wilson, 1994; Zumbo,
1999). Measurement instruments with a high degree of differential item
functioning (DIF) may not assess what it is intended to assess and a sign
of low construct validity (Kristensen, Bjorner, Christensen, & Borg,
2004). In traditional psychometric methods for validation of items from
multi-item scales (e.g., Cronbachs alpha, factor analysis, and inter-item
correlations), focus is on evaluating how well each item assesses the
concept that is related to the instrument (Petersen et al., 2003). One way
to do this is by calculating the internal consistency level which describes
the relationship between items in scale and subscale, respectively. An
acceptable Cronbachs alpha (internal consistency) does not exclude the
possibility of severe item bias, because the Cronbachs alpha does not
consider exogenous variables such as e.g., gender, age or occupation
(Groenvold, Bjorner, Klee, & Kreiner, 1995). According to Groenvold et
al. (1995) this emphasizes the point that traditional psychometric tests like
Cronbachs alpha, factor analysis, and inter-item correlations fail to take
analytic purposes on an index into account. Cronbachs alpha is calculated
in a context that is detached from the clinical problem investigated.
Therefore, it is of limited interest if the internal consistency is excellent, if
the index is severely biased in relation to a subgroup used for grouping in
the analyses. This is an interesting aspect which extends the model of
construct validation to include the intentional use of the instrument.

32
Statements like this instrument is known to have good validity and
reliability lose their meaning, because the instrument performed well in
another population or with other research questions, and does not exclude
the possibility that the use of this instrument in new contexts will result in
severe information bias. This knowledge is not new (Nunnally &
Bernstein, 1994), but so far not applied in the traditional procedures used
for construct validation in work stress research.

Exploratory factor analysis


The idea of EFA is to find meaningful relations between items, or
meaningful clusters of items, usually called factors. Factor analysis is a
rather subjective enterprise. In both EFA and CFA there are a number of
different choices for the researcher to make, and these choices can have a
big influence on the outcome of the analysis (Nunnally & Bernstein,
1994). For example, the researcher has to choose which method that
should be used, the number of factors extracted, if the factors should be
rotated, and last, but not least, interpret the results in terms of meaning.
The number of factors to be retained is a critical decision in EFA, and
there are several decision rules here (see e.g., Glorfeld, 1995, for a review).
The minimum criterion is that factors with eigenvalue greater than one
are extracted (Kaiser, 1960). Since this method has a tendency to extract
too many factors, Cattells (1966) scree test method is often used as a
diagnostic tool. Both of these methods have advantages, but perhaps the
most useful decision rule is to extract the number of factors that produce
the solution that makes the best theoretical sense, provided that other
statistical decision rules have been taken into account.
The question of rotation of extracted factors is another essential part
of the EFA process. There are two different rotation techniques, those
which assume that factors are uncorrelated (orthogonal rotational
techniques) and those which allow factors to correlate (oblique rotational
techniques). In Study I the factors are highly correlated, and therefore an
oblique promax rotation was used.

Differential item functioning


There are several methods existing for evaluation and identification of
Differential item functioning (DIF, c.f. Kristjansson et al., 2005). In the
present study ordinal logistic regression is used to evaluate DIF in JSS
items. A reason for selecting this method was that it is considered to be
more general and flexible than the other DIF procedures (Swaminathan &
Rogers, 1990). Furthermore, ordinal logistic regression is a suitable

33
method for detecting DIF in ordinal items (Kristjansson et al., 2005). DIF
is based on the ordinal logistic regression equation

y = b0 + b1TOTAL + b2GRP + b3TOTAL GRPi + i

Zumbos (1999) DIF concept measures the effects of group and the
interaction over and above the total scale score. The test for statistical
significance follows a hierarchy of steps for entering variables into the
model. In the first step the total score variable is entered, at the second
step, the grouping variable (e.g., gender) is included. In the last step, the
interaction term between the variables included in the first and second
stage entered. The last step in the analysis describes whether the difference
between the group responses on an item changes over the latent variable
continuum.
With the results from the chi-squared test for logistic regression from
the first and third step, one can calculate the significant level through
subtracting the chi-squared value in step three from the value in the first
step. The differences in chi-squared value can then be compared to its
distribution function with 2 degrees of freedom (3 - 1 = 2 df). To
calculate the R-squared level, the same procedure is used as for chi-squared
calculation.
For an item to be classified as displaying DIF, two criteria must be
met: First, the chi-squared must have a p value less or equal to .01.
Secondly, the effect size measure must have an R-squared value of at least
.035. In my thesis I use the effect-size criteria suggested by Jodoin & Gierl
(2001) to quantify the magnitude of DIF, that is, DIF is negligible for
effect-size values below .035, moderate for levels between .035 and .070,
and large for levels above .070.

34
SUMMARY OF STUDIES
This section briefly summarizes the results from the three empirical papers
that constitute the basis of this thesis. The first study assesses and evaluates
the psychometric properties of a Swedish version of the Job Stress Survey
(JSS). The second study examines JSS with special attention to the
problems of differential item functioning on subgroups of gender and
occupation. In the third study a longitudinal design were used in order to
study the relationship between workplace stress and health over a four-
year period, another purpose with the third study were to investigate the
ability of JSS to relate to health, especially musculoskeletal problems.

Study I
Holmstrm S., Molander, B., Jansson, J., & Barnekow-Bergkvist, M.
(2008). Evaluation of a Swedish version of Job Stress Survey.
Scandinavian Journal of Psychology, 49, 277-286.

The main purpose of the present study was to evaluate a Swedish version
of JSS (Spielberger & Vagg, 1999). In particular, we focused on the
psychometric properties of the instrument and to what extent it would be
possible to verify the original factor structure in this version. There are
relatively little published data on the psychometric properties of the JSS,
which makes it hard to fully assess the strength of the instrument (Rick et
al., 2001). However, the work that has been published so far is promising.
It is important to make a careful and thorough evaluation of the original
instrument to find out how it works in different settings and in various
occupational groups. An important part of the process of establishing
cross-cultural construct validity is to calculate and evaluate the
psychometric properties of the adaptations of the instrument in various
countries.
In this study 1186 employees participated, working in metal assembly
industry and in medical service in northern Sweden. The age of the
employees varied between 17.5 and 64.9 years (M age = 43.1, SD = 11.1),
the gender distribution were 51.3 % men, and the response rate for the
whole sample was 79.0 %. The stress in this study was laid on replication
rather than theory testing, and therefore exploratory factor analyses were
used to identify the underlying factor structure of the JSS.
The initial analysis indicated that a five-factor model was best for
describing the item intercorrelations in the Severity scale. A scree plot
analysis showed that a two-factor model was the best solution, with the
subscales Lack of Organizational Support and Job Pressure. This model is

35
in line with the findings reported by Spielberger and Vagg (1999), who
found two strong factors and two or three weaker factors for both men
and women. The same factor structure was found for both the Frequency
and Index scales.
There are some differences between men and women in the Severity
scale with respect to how items load on the different factors. There are
four items that have different loading patterns for men and women.
Interestingly, for men, the four items load on the same factors as in the
original instrument, whereas for women these items load on the opposite
factors. In the American version there was a difference in loading pattern
between men and women too, but if this difference is a real gender
difference in JSS, an effect of difference in work setting (e.g., duties,
responsibilities, organizational level), or due to cultural differences is hard
to assess.
Occupational level seems to have a stronger impact on job stress than
gender (Vagg & Spielberger, 1998), because the interaction with
organizational level influences the reaction of men and women to different
sources of stress (Vagg et al., 2002). To evaluate gender differences in JSS,
further analysis on item level is therefore needed.
The results also showed that the same factor structure was found to be
stable for both the metal industry and the medical sectors in the analyses
of the Severity, Frequency, and Index scales. When metal assembly
workers and medical service workers were tested separately, the same two
subscales, Job Pressure and Lack of Organizational Support, were found in
the Severity, Frequency, and Index scales.
Furthermore, the results from the factor analyses showed a good fit
between the Swedish version and the American version with item loadings
over .4, or greater, on both subfactors in all three scales (Severity,
Frequency, and Index). These results confirm previous analyses of the
factor structure (Spielberger, 1991; Spielberger & Reheiser, 1994a;
Spielberger, Reheiser, Reheiser, & Vagg, 2000; Vagg & Spielberger,
1998). Analyses of the internal consistencies of the different scales revealed
alpha levels between .84 and .95, which are above Nunnally and Bersteins
(1994) cut off level of .70. The values reported in this study were in line
with the levels reported for the American version (Spielberger & Reheiser,
1994a; Spielberger & Vagg, 1999). Test-retest reliability showed on all
scales a large and positive correlation between the first and second
measure. The concurrent validity for JSS was evaluated by using the PSQ
instrument (Bergdahl & Bergdahl, 2002; Levenstein et al., 1993) as
comparison, and the pattern of correlations between the JSS and PSQ
indicated good concurrent validity in the JSS instrument.

36
Concerns have been raised over the similarity in the factor structure
between severity and frequency. These variables are meant to measure
different aspects of stress (Kasl, 1998; Vagg & Spielberger, 1998). The
similarities in factor structure raise questions about the independence of
the severity and frequency variables. Kasl (1998) argues that substantial
correlations between Severity and Frequency would undermine the claim
that Frequency yields new information. However, the results from the
present study show that correlations between Severity and Frequency are
only low to moderate. On the other hand, results from the test-retest
study showed a significant decline in mean value between the first and
second measurement for the Severity and Frequency scales. This finding
indicated that ratings of severity and frequency were not completely
detached from each other. Another finding that may be taken as evidence
for the importance of frequency was that frequency had stronger impact
than severity on the different index scales.
In summary, the overall conclusion of this study was that the present
version of JSS shows a stable factor structure, satisfactory reliability, and
high concurrent validity.

Study II
Holmstrm, S., Molander, B., & Barnekow-Bergkvist, M. (2007).
Gender, occupation, and item bias: Job stress assessed by Job Stress
Survey. Manuscript submitted for publication.

When evaluating self-report instruments care must be taken so that


measurements are appropriate for different important subgroups (e.g.,
gender, age, occupation, occupational level). If there are differences in
response patterns between subgroups, this can be a sign of item bias
known as differential item functioning (Smith, 2002; Swaminathan &
Rogers, 1990; Zumbo, 1999). Differential item functioning (DIF) exists if
one or more item is more difficult, discriminating, or easily guessed for
one group than for another. For example, all persons at a given stress level
should have the same probability of endorsing an item in the same way
regardless of group membership (Osterlind, 1983).
In this study the main purpose was to analyse the Swedish version of
JSS (Holmstrm et al., 2004; Holmstrm et al., 2008) with respect to the
problem of differential item functioning, and with special attention to the
subgroups of gender and occupation (assembly industry workers and
medical service personnel). Perceived severity, frequency, and overall
occupational stress for subgroups of gender and occupation (metal
industry workers and medical service personnel) were analysed as well.

37
The participants were 1186 employees working in metal assembly
industry and in medical service in northern Sweden. Analyses for DIF
were conducted in two stages. First, gender and occupation in the whole
sample were examined. The two subgroups showed an even distribution
between men and women in the gender subgroup. Second, gender was
examined in each of the two occupational groups, that is, in the medical
service and assembly industry groups. The gender breakdown within the
occupational subgroups showed an imbalance, that is, the percentage of
men in the industry subgroup was 78.8 %, and the percentage of women
in the medical subgroup was 77.5 %.
In the analyses of gender performed on the whole sample there were
no items that met the criteria for DIF, neither in the main scales, nor in
the subscales. In the analyses of occupation, on the other hand, DIF was
found in a total of eight items. Five of these items appear in frequency
scales and three of these five items also appear in index scales.
It is noteworthy that there was no item in the Severity dimension that
exceeded the criteria for DIF. Most of the items that showed DIF were
found in the Frequency dimension, which may be a sign of the differences
in the structure of the organization and work settings between metal
industry and medical service.
The MANOVAs on item level showed significant multivariate effects
for both the Severity and Frequency dimension of JSS for gender,
occupation, and for the interaction between gender and occupation.
Women have generally higher mean values than men in the Severity and
Frequency items, and the ANOVA post-hoc tests showed that women had
significantly higher mean value ratings in fourteen Severity items and in
three Frequency items.
In the occupational subgroup, post-hoc tests showed that medical
service personnel had higher mean values than assembly industry
personnel on twenty-three of twenty-nine Severity items, and seven of
them were significant. Metal-industry personnel had seven items with
higher mean value, and three of them were significant. In the Frequency
dimension there were twelve items that showed significant differences,
eleven of which with higher mean values for medical service personnel.
The same gender trend as at the item level was noticeable at the scale
level, and there were significant main effects for gender on five of the nine
scales. There were deviations in mean values between different
occupations. Medical service personal had higher mean values on seven of
the JSS scales and subscales, and three of the scales showed a significant
difference.

38
The main findings of this study were that both gender and occupation
had a substantial impact on specific sources of occupational stress as
assessed by JSS scales and individual items. The results support the
conclusion that the differences in response pattern for the actual items
depend on differences in work settings in the two occupational groups.
These results strengthen previous findings concerning gender effects of
stress in work places, and suggest that increased attention should be
directed to DIF analyses in studies on differences in perceived stress
between occupational groups and branches of occupations.

Study III
Holmstrm, S., Molander, B., & Barnekow-Bergkvist, M. (2008). Work-
related stress and musculoskeletal problems in metal industry: A
longitudinal study. Manuscript.

The main purpose of this study was to examine the longitudinally


relationships between different types of self-reported stress and health,
especially the relationships between stress and musculoskeletal complaints.
Another aim was to find out how lack of social support relates to
musculoskeletal complaints. A further purpose was to examine the present
version of JSS (Holmstrm et al., 2008; Spielberger & Vagg, 1999) with
respect to health.
It is well known that high level of stress is associated to
musculoskeletal complaints and pain (e.g., Lundberg et al., 1994; Siivola
et al., 2004). Risk factors for musculoskeletal diseases are physical,
psychosocial, or individual factors (Arins et al., 2001). Reviews of the
physical and psychosocial factors (e.g., Arins et al., 2000, 2001;
Hoogendoorn et al., 2000; Larsson et al., 2007; van der Windt et al.,
2000) show that physical factors, such as repetitive movements, awkward
postures, high force demands, work posture, vibration are important, as
well as, psychosocial factors, such as low co-worker support, high
quantitative demands, low job control and low job satisfaction. These
factors might be associated with musculoskeletal disorders directly, or as a
consequence of raising the level of stress in the work place.
Social support has been found to reduce risk for different problems,
such as back pain (e.g., Clays et al., 2007; Hartvigsen et al., 2004;
Hoogendoorn et al., 2000), pain in neck and shoulder (e.g., Arins et al.,
2001; Bongers et al., 2006; Larsson et al., 2007) or musculoskeletal
problems in general (e.g., Woods, 2005). Social support is not often seen
as a factor of stress, and through the use of the JSS instrument we were
given an opportunity to compare the relationship between health and the

39
two subscales Lack of Organizational Support and Job Pressure. A further
advantage of the JSS instrument was that it made it possible to consider
both the severity and the frequency of the stressors.
The study was performed on participants from two comparable metal
and assembly factories located in the same town in northern Sweden. Data
were collected at four points, the first in November 2001, the second six
months later, and then every 12th month. All employees at the two
factories were asked to participate in the study, at time T1 there were 73
people employed at factory A, and a response rate of 74.0 %. Factory B
had 70 people employed and the response rate was 87.1 %. The
longitudinal sample consists of 56 employees who participated in all four
measures, and they were evenly distributed within the two factories.
The level of stress was high at both factories, and compared with
Swedish norms for metal industry work-places (Holmstrm et al., 2008;
Molander, Holmstrm, & Jansson, 2007) the level of stress measured by
the JS-X scale was well above average, and close to the 70th percentile.
This means that employees at both factories may have experienced levels
of stress which are likely to have severely affected their functioning at the
workplace, and might also have been a serious threat to their health.
MANOVA was used when comparing the two factories for work-related
stress and subjective health, the results showed no significant differences.
This means that both factories have similar levels of work-related stress
and health at the four measurement points. The 2 (Factory) x 4 (Time of
Measurement) ANOVAs were done to analyse stress and health in the
longitudinal sample, the result showing significant main effects for JS-S,
JP-S, and for a psychosocial composite measure (ps < .03). Also, for both
JSS scales there were significant Factory x Time of measurement
interactions (ps < .005). For factory B, the interaction indicates a
decreasing trend of severity ratings over time, whereas factory A shows a
trend of increasing ratings of severity.
Figure 1 illustrates how the scores for the overall job stress scales (JS-
X, JS-S, and JS-F) chang over the four measurement occasions and
between factories for the longitudinal sample of 56 employees as
compared to the other participating employees in each factory. It should
be noted in Figure 1 that both factories show similar up-and- down
patterns with respect to level of stress over the four measurement points.
These patterns, which were especially pronounced in the longitudinal
sample and for the scales measuring job pressure, are likely to mirror the
volume of production in the factories. Unfortunately, objective measures
of trade cycle and volume of production were not possible to obtain, but
one item in the JSS scales which could be seen as especially associated to

40
such measures was Working overtime. The mean ratings of this item
matched the changes in Figure 1 very well.

Figure 1. Mean ratings of job stress (JS-X. JS-S. JS-F) as a function of


factory and measurement occasion in employees participating
longitudinally (left panel) and remaining employees participating cross-
sectionally (right panel) (Study III, Holmstrm et al., 2008).

Mean ratings for musculoskeletal complaints in factory B were higher


compared to ratings in factory A, although the latter employees reported
higher level of stress. The mean rating levels for back pain, neck pain and
pain in shoulders are for both factories not alarmingly high, but they were
higher than the mean ratings obtained in a representative sample of other
metal industry work sites, and notably higher than what was obtained in
the medical sector and in a representative sample of Ume citizens
(Molander et al., 2007). Results from the present study confirm previous
studies (e.g., Statistics Sweden, 2006) that back and shoulders are the two

41
categories with most musculoskeletal problems and that legs and wrists are
the two categories with least pain problems.
Results showed quite clearly that LS scales were more associated with
pain than JP scales, and these results add to the bulk of findings showing
that low social support is an important factor for understanding
musculoskeletal pain (e.g., Clays et al., 2007; Hartvigsen et al., 2004;
Hoogendoorn et al., 2000). In the present study high correlations were
obtained between LS-scales and musculoskeletal complaints for back, neck
and shoulders, but also for arms, wrists, and legs. The lack of studies
showing effects on lower parts of the body may simply mean that
industrial workplaces are underrepresented in social-support studies. The
association between Severity and Frequency ratings respectively, and
health ratings showed that LS-S scale produced both more significant and
higher values on correlations than the LS-F scale. When only selecting LS-
items of direct individual concern, correlation levels increased in size, as
expected, but only for the LS-S scale. Both the LS-X scale and the LS-F
scale showed reduction. This result again demonstrates the difference
between severity ratings and frequency ratings, but also points to the
importance of keeping different categories of social support in mind when
discussing the relationship to musculoskeletal complaints. In reviews of
the effects of low social support various categories of support are not
always kept apart (c.f. Kristensen, 1995; Woods, 2005).
This study revealed that the variation in the different measures
between measurement occasions was larger and more dynamic than
expected. It could be that seemingly small changes in production cause
quite large changes in the stress and musculoskeletal measures. Some
support for this reasoning comes from a comparison between the two
factories.
As this study showed, even if the factories are similar with respect to
location, cultural values, type of production, structure, and size, such
similarities may still be misleading regarding assumptions about equal
terms with respect to relationships among stress, musculoskeletal and
psychosocial measures. This observation points to the potential problem
with scientific reviews, where results from many similar factories are
averaged, and the problem of selecting suitable control factories in
intervention studies.
An important result is that the JSS instrument is well suited for studies
of relationships between work-related stress and complaints related to
health in the metal industry environment. The relationships with
musculoskeletal ratings as well as with ratings of other health problems are
strong and significant. JSS scales evaluating lack of social and

42
organizational support were shown to be more important for the stress-
musculoskeletal complaints relationship than scales directed to job
pressure.

43
GENERAL DISCUSSION
The overall objective of the thesis was to make an evaluation of the Job
Stress Survey (Spielberger, 1991; Spielberger & Vagg, 1999). Factor
analysis from Study I showed good resemblance between the Swedish
version and the American version by providing the same two subscales
(Job Pressure and Lack of Organizational Support) for the Severity,
Frequency, and Index scales. When analysed on occupational and gender
level, similar factor structures were found for both metal industry and the
medical sectors, and for men and women. These results confirm previous
analyses of the factor structure (Spielberger, 1991; Spielberger & Reheiser,
1994a; Spielberger et al., 2000; Spielberger & Vagg, 1999; Vagg &
Spielberger, 1998).
The reliability tests showed that alpha values of internal consistency
for different subgroups in the three studies are in line with levels reported
by the American version (Spielberger & Reheiser, 1994a; Spielberger &
Vagg, 1999). A notable finding from Study I was that the alpha values on
the Job Stress scales (Severity, Frequency, and Index) are all above .90.
This raises the question of redundancy between some items in the
instrument, and that a reduction of the JSS instrument could capture the
same underlying features (Helms, Henze, Sass, & Mifsud, 2006). The JSS
instrument consists of 30 items, and a reduction of the number of items
would give a smaller alpha level on the actual scales, but a reduction in
number of items would also mean a loss in variety of work settings.
The test-retest reliability was evaluated and showed high levels for the
different scales. Stability coefficients over the different measurement
occasions were calculated for the participants included in the longitudinal
sample. Stability measures were somewhat low but acceptable (c.f.
Loewenthal, 1996). The Swedish version also showed good concurrent
validity in comparison with the PSQ (Bergdahl & Bergdahl, 2002;
Levenstein et al., 1993).
In Study II, analyses of items demonstrate that both gender and
occupation has a substantial impact on specific sources of occupational
stress assessed by JSS scales, subscales and the individual items. Thus, it
was of importance to assess if the observed differences were due to item
bias. Analyses of DIF were carried out with the Ordinal Logistic
Regression method (Kristjansson et al., 2005; Zumbo, 1999). The overall
results of DIF analyses showed no item bias for the gender subgroup, but
for the occupational subgroup. There were a total of eight items which
exceeded magnitudes according to both DIF criterias. Six of the items
labelled with DIF for the occupational subgroup were found in the

44
Frequency (JS-F) and Index (JS-X) scales. There is no clear-cut level for
how many items with DIF a scale can handle without being biased for a
group (Zumbo, 1999). In this case the overall results show that
conclusions made on scale level are not jeopardized because of the DIF
items, since they are too few to make an impact on the overall result on
the different scales. When comparing or analysing single items on
occupational level, caution is needed so that comparisons and conclusions
is not based on items showing DIF bias. It is therefore crucial to make
DIF analyses to identify biased items in important subgroups, and
conducting analyses of DIF should be routine when comparing groups,
but also when instruments are developed and translated. If this is done the
quality and the precision will increase, and most importantly, any
implications drawn will be more correct.
The findings in Study II support the conclusion that the differences in
response pattern for the actual items depend on variations in work settings
between the two occupational groups. One such difference of possible
importance is that one group has more of a person-to-person interaction
(medical service personnel) and the other group more of a person-to-
machine interaction (metal industry personnel). The results of the analyses
on scale and item level showed numerous gender differences in the severity
and frequency of occurrence of specific JSS stressors. The findings in
Study II confirms earlier studies in that women in general report more
stress (Narayanan et al., 1999; Nolan-Hoeksema, 1990; Spielberger &
Reheiser, 1994a; Spielberger & Vagg, 1999; Vagg et al., 2002), more
interpersonal stressors than men (Treaster & Burr, 2004), and have higher
levels on both the Severity and on the Frequency items (Krajewski &
Goffin, 2005; Narayanan et al., 1999; Spielberger & Reheiser, 1994b;
Spielberger & Vagg, 1999; Triosi, 2001) compared to men. The results
from Study II shows that there are differences in both gender and
occupation on both scale and item level, which indicates that it is
important to develop separate norms for gender. This has also been done
in the original JSS version (Spielberger & Vagg, (1999) as well as in the
development of the Swedish version (Holmstrm et al., 2008). Similarly,
there is need for developing separate norms for specific work groups that
perform similar duties and tasks, and for occupational level. This would
help interpretation of gender differences.
Another purpose with this thesis was to evaluate the JSS instrument
with respect to relationships with musculoskeletal problems and several
other psychosocially related problems, such as stomach pain, tiredness,
and worry. The result from Study III showed that the JSS instrument is
well suited for studies of relationships between work-related stress and

45
complaints related to health in the metal industry environment. The
relationships with musculoskeletal ratings as well as with ratings of other
health problems were differentiated and significant. However,
musculoskeletal and psychosocial variables seem to be better predictors
then the stress scales when to predict musculoskeletal complains later on.
The findings points to the fact that stress do not have a direct effect on
musculoskeletal complaints, but mediating roll.
It was found quite clearly that support scales (LS) were more
associated with pain than pressure scales (JP). These results add to the
bulk of earlier findings showing that low social support is an important
factor for understanding musculoskeletal pain (e.g., Clays et al., 2007;
Hartvigsen et al., 2004; Hoogendoorn et al., 2000). Most findings
reported in the literature on the relation between support and
musculoskeletal pain related to pain in neck, back pain, and pain in
shoulders. There are only a few studies which report associations to other
body parts such as wrists, arms, and hands. Even rarer are reports of effects
for lower parts of the body. Employees in the metal industry group in
Study III showed high correlations for back, neck and shoulders, but also
for arms, wrists, and legs. The lack of studies on lower parts of the body
may simply mean that industrial workplaces are underrepresented in
social-support studies and reviews. Social support components can be
defined and categorised in different ways (e.g., Berkman, Glass, Brissette,
& Seeman, 2000; Cohen, 2004; Woods, 2005). When making
calculations using only LS-items of direct individual concern, the
correlation coefficients increased in size, as expected, but only for the LS-S
scale. Both the LS-X index scale and the LS-F scale showed reduced sizes.
This shows the importance of keeping different categories of social
support in mind when discussing the relationship to musculoskeletal
complaints.
The psychometric quality of the Swedish version of the JSS
instrument is high, as shown by this thesis. Results from Study III showed
that JSS is useful when examining relations between stress and health. Kasl
(1998) have criticized JSS for lacking in questions dealing with
monotonous jobs and person-to-machine interactions. The main purpose
with the JSS is to measure generic sources of occupational stress
encountered in a wide variety of work settings. The gain of using generic
stress instrument is that the result can be compared with other
departments or organizations. However, the result from Study III points
out that it will be useful to complement the JSS instrument with some
instrument for measuring stress caused by the manufacturing itself that is,
different products or services may involve different levels of trouble along

46
the line. Stress data closer to the production could help increase the
understanding of the sometimes fast-changing stress-musculoskeletal
relationships.

Severity and frequency ratings of stressors


When developing the JSS instrument Spielberger and colleagues followed
the recommendation of DeFrank (1988) and Dewe (1989) to give more
attention to both the perceived severity of the stressor, and the frequency
of the stressor. Both of these variables need to be taken into account to get
a more complete understanding of the stress experience in different
occupations or groups of employees, and it appears that both perspectives
are equally important for evaluating occupational stress (Vagg &
Spielberger, 1998). According to Spielberger (1983), the difference
between perceived severity of work-related stressors and how often they
are experienced is comparable to distinguishing between emotional states
and traits. Motowidlo, Packard, and Manning (1986) noted in their study
that stressful events appear to be functions of both external work
conditions and personal characteristics, while perceived severity is
influenced primarily by the latter. In Kasls (1998) review of the JSS some
concern was raised over the independency of the severity and frequency
variables. Kasl argues that the similarity in the factor structure between
Severity and Frequency undermine the claim that frequency is an
independent variable that yields new information. He points out that
substantial correlations between Severity and Frequency scales would
undermine this claim. Findings in Study I showed moderate levels of
correlation between Severity and Frequency scales. Result from the test-
retest in Study III showed a significant decline in mean value between the
first and second measurement for the JS-S and JS-F scales. This finding
indicates that severity and frequency ratings are not completely detached
from each other. The findings in Study III and in Spielberger and Vagg
(1999) showed that the Frequency scales had bigger impact than the
Severity scales on the total Index scales. The result in Study III showed
that severity ratings are more important than frequency ratings for the LS
relationships with musculoskeletal pain. The LS-S scale produced more
significant correlations and higher values than the LS-F scale. The lesser
importance of LS-F in the context of associations with musculoskeletal
ratings may simply be due to the focusing on present situation in the latter
ratings, and an addition of rated duration of pain to such measures could
change the picture.

47
Further evaluation and development of the JSS
It is important to make a careful and thorough evaluation of the original
instrument to find out how it works in different settings and in various
occupational groups. An important part of the process of establishing
cross-cultural construct validity is to calculate and evaluate the
psychometric properties of the adaptations of the instrument in various
countries. After that, changes can be made and new versions be developed.
The standard stressor Assignment of disagreeable duties, the item
which the other twenty-nine Severity items were compared to, had in
previous research been given an average rating by participants from
different occupations (Grier, 1982; Spielberger et al., 1981), and therefore
this stressor is given a fixed midpoint value of 5 in the Severity scale. In
the Spielberg and Reheiser (1994b) study, the ranks for the standard
stressor item were thirteen and sixteen for women and men, respectively.
Spielberg and Reheiser interpret this result as additional evidence that the
standard stressor is about average in perceived severity as compared to the
other JSS items. However, the conclusion that Spielberg and Reheiser
make is too swift, because the respondents have not actually made any
assessment of this stressor and comparisons with other items are therefore
not especially meaningful. Hodapp, Tanzer, Korunka, Maier, & Pestemer
(2005), showed in their study that the German translation of the standard
stressor did not evoke the same amount of stress for all respondents. This
is a cross-cultural problem, which should be given more attention in the
future.
More attention should also be given to the different ratings of Severity
and Frequency. In the original version of JSS, Severity and Frequency
ratings are combined into the Index scale. One important issue to evaluate
in the future is when different combinations of Severity and Frequency
ratings results in the same levels of stress in the Index scale. For example, if
one participant rates 3 on a Severity item and 3 on a Frequency item, then
the Index will be 9. On the same item, another participant makes ratings
of 9 and 1 respectively, and a third participant makes a rating of 1 and 9.
For all three items the score on the Index scale will be 9, however, it may
be unlikely that the participants perceive stress in the same way. The
present version of the test makes no such distinction, and in the future
more attention should be given to evaluate whether Severity and
Frequency ratings should have the same influence on the Index scale.

48
REFERENCES
Arins, G. A. M., van Mechelen, W., Bongers, P. M., Bouter, L. M., &
van der Wal, G. (2000). Physical risk factors for neck pain.
Scandinavian Journal of Work Environment Health, 26, 7-19.
Arins, G. A. M., van Mechelen, W., Bongers, P. M., Bouter, L. M., &
van der Wal, G. (2001). Psychosocial risk factors for neck pain: A
systematic review. American Journal of Industrial Medicine, 39, 180
193.
Beehr, T. A. (1995). Psychological stress in the workplace. London:
Routledge.
Beehr, T. A., & Newman, J. E. (1978). Job stress, employee health and
organizational effectiveness: A facet analysis, model and literature
review. Personnel Psychology, 31, 665-699.
Bergdahl, J., & Bergdahl, M. (2002). Perceived stress in adults: Prevalence
and association of depression, anxiety and medication in a Swedish
population. Stress and Health, 18, 235-241.
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From
social integration to health: Durkheim in the new millennium.
Social Science & Medicine, 51, 843-857.
Bhagat, R. S., Allie, S. M., & Ford, J. D. L. (1995). Coping with stressful
life events: An empirical analysis. In R. Crandall & P. L. Perrew
(Eds.), Occupational stress. A handbook (pp. 93-112). Washington:
Taylor & Francis.
Blair, S., Djupsjbacka, M., Johansson, H., Ljubisavljevic, M., Passatore,
M., Punnett, L., & Windhorst, U. (2003). Neuromuscular
mechanisms behind chronic work-related myalgias: An overview. In
H. Johansson, U. Windhorst, M. Djupsjbacka & M. Passatore
(Eds.), Chronic work-related myalgia (pp. 5-47). Gvle: Gvle
University Press.
Bongard, S., & alAbsi, M. (2005). Domain-specific anger expression and
blood pressure in an occupational setting. Journal of Psychosomatic
Research, 58, 43-49.
Bongers, P. M., Ijmker, S., van den Heuvel, S., & Blatter, B. M. (2006).
Epidemology of work related neck and upper limb problems:
Psychosocial and personal risk factors (Part I) and effective
interventions from a bio behavioural perspective (Part II). Journal of
Occupational Rehabilitation, 16, 279-302.

49
Bongers, P. M., Kremer, A. M., & ter Laak, J. (2002). Are psychosocial
factors, risk factors for symptoms and signs of the shoulder, elbow or
hand/wrist?: A review of the epidemiological literature. American
Journal of Industrial Medicine, 41, 315-342.
Bongers, P. M., Winter, C. R., & Kompier, M. A. J. (1993). Psychosocial
factors at work and musculoskeletal disease. Scandinavian Journal of
Work, Environment and Health, 19, 297-312.
Bosma, H., Marmot, M. G., Hemingway, H., Nicholson, A. C., Brunner,
E., & Stansfeld, S. A. (1997). Low job control and risk of coronary
heart disease in Whitehall II (prospective cohort) study. British
Medical Journal, 314, 558-565.
Bosma, H., Peter, R., Siegrist, J., & Marmot, M. G. (1998). Two
alternative job stress models and the risk of coronary heart disease.
American Journal of Public Health, 88, 68-74.
Bourbonnais, R., Comeau, M., & Vezina, M. (1999). Job strain and
evolution of mental health among nurses. Journal of Occupational
Health Psychology, 4, 95-107.
Brief, A. P., & George, J. M. (1991). Psychological stress in the
workplace: A brief comment on Lazarus outlook. In P. L. Perrw
(Ed.), Handbook on job stress (pp. 15-20). Corte Madera, CA: Select
Press.
Calnan, M., Wainwright, D., & Almond, S. (2000). Job strain, effort-
reward imbalance, and mental distress: A study of occupations in
general medical practice. Work & Stress, 14, 297-311.
Cannon, W. (1914). The interrelations of emotions as suggested by recent
physiological researches. American Journal of Psychology, 25, 256-
282.
Cannon, W. (1929). Bodily changes in pain, hunger, fear, and rage. Boston,
MA: C.T. Branford Company.
Cannon, W. (1935). Stresses and strains of homeostasis. American Journal
of Medical Science, 189, 1-14.
Caplan, R. D. (1987). Person-environment fit theory and organizations:
Commensurate dimensions, time perspectives, and mechanisms.
Journal of Vocational Behavior, 31, 248-267.
Cartwright, S., & Cooper, C. L. (1997). Managing workplace stress.
Thousand Oaks, CA: Sage.
Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F.,
Robinson, D. S., Ketcham, A. S., Moffat, F., & Clark, K. C. Jnr.
(1993). How coping mediates the effect of optimism and distress: a
study of women with early stage breast cancer. Journal of Personality
and Social Psychology, 65, 375-390.

50
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing
coping strategies: A theoretically based approach. Journal of
Personality and Social Psychology, 56, 267-283.
Cattell, R. B. (1966). The meaning and strategic use of factor analysis. In
R. B. Cattell (Ed.), Handbook of multivariate experimental psychology
(pp. 174-243). Skokie, IL: Rand NcNally.
Chemers, M. M., Hays, R. B., Rhodewalt, F., & Wysocki, J. (1985). A
person-environment analysis of job stress: A contingency model
explanation. Journal of Personality and Social Psychology, 49, 628-
635.
Cheng, Y., Kawachi, I., Coakley, E. H., Schwartz, J., & Colditz, G.
(2000). Association between psychosocial work characteristics and
health functioning in American women: Prospective study. British
Medical Journal, 320, 1432-1436.
Clays, E., De Bacquer, D., Leynen, F., Kornitzer, M., Kittel, F., & De
Backer, G. (2007). The impact of psychosocial factors on low back
pain. SPINE, 32, 262-268.
Cohen, S. (2004). Social relationships and health. American Psychologist,
59, 676-684.
Cole, S. R. (1999). Assessment of differential item functioning in the
Perceived Stress Scale-10. Journal of Epidemiology and Community
Health, 53, 319-320.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design
and analysis for field settings. Boston: Houghton Mifflin.
Cooper, C. L. (1986). Job distress: Recent research and the emerging role
of the clinical occupational psychologist. Bulletin of the British
Psychological Society, 39, 325-331.
Cooper, C. L. (1987). The experience and management of stress: Job and
organizational determinants. In A. W. Riley & S. J. Zaccaro (Eds.),
Occupational stress and organizational effectiveness (pp. 53-69). New
York: Praeger.
Cooper, C. L., & Cartwright, S. (1994). Healthy mind: Healthy
organization - a proactive approach to occupational stress. Human
Relations, 47, 455-470.
Cooper, C. L., & Dewe, P. J. (2004). Stress: A brief history. Oxford:
Blackwell Publishing.
Cooper, C. L., Dewe, P. J., & Driscoll, M. P. (2001). Organizational
stress: A review and critique of theory, research and applications.
Thousand Oaks, CA: Sage.
Cooper, C. L., Sloan, S. J., & Williams, J. (1988). Occupational Stress
Indicator Management Guide. Oxford, England: NFER-Nelson.

51
Cox, T. (1978). Stress. London: MacMillan.
Dallner, M., Elo, A.-L., Gamberale, F., Hottinen, V., Knardahl, S.,
Lindstrm, K., Skogstad, A., & rhede, E. (2000). Validation of the
General Nordic Questionnaire, QPSNordic, for psychological and social
factors at work. Nordic Council of Ministers, Copenhagen: Nord
2000:12.
Davis, K. G., & Heaney, C. A. (2000). The relationship between
psychosocial work characteristics and low back pain: Underlying
methodological issues. Clinical Biomechanics, 15, 389-406.
Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work
adjustment. Minneapolis: University of Minnesota Press.
Day, A. L., & Livingstone, H. A. (2001). Chronic and acute stressors
among military personnel: Do coping styles buffer their negative
impact on health? Journal of Occupational Health Psychology, 6, 348-
360.
de Jonge, J., Bosma, H., Peter, R., & Siegrist, J. (2000). Job strain, effort-
reward imbalance and employee well-being: A large-scale cross-
sectional study. Social Science & Medicine, 50, 1317-1327.
de Jonge, J., & Kompier, M. A. J. (1997). A critical examination of the
demand-control-support model from a work psychological
perspective. International Journal of Stress Management, 4, 235-258.
de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I. L. D., &
Bongers, P. M. (2003). "The very best of the millennium":
Longitudinal research and the demand-control-(support) model.
Journal of Occupational Health Psychology, 8, 282-305.
De Raeve, L., Jansen, N. W. H., & Kant, I. (2007). Health effects of
transitions in work schedule, workhours and overtime in a
prospective cohort study. Scandinavian Journal of Work,
Environment & Health, 33, 105-113.
Dean, K., Holst, E., Kreiner, S., Schoenborn, C., & Wilson, R. (1994).
Measurement issues in research on social support and health. Journal
of Epidemiology and Community Health, 48, 201-206.
DeFrank, R. S. (1988). Psychometric measurement of occupational stress:
Current concerns and future directions. In J. J. Hurrell, L. R.
Murphy, S. L. Santer & C. L. Cooper (Eds.), Occupational stress:
Issues and developments in research (pp. 54-65). New York: Taylor &
Francis.
Dewe, P. J. (1989). Examining the nature of work stress: Individual
evaluations of stressful experiences and coping. Human Relations,
42, 993-1013.

52
Dewe, P. J., Cox, T., & Ferguson, E. (1993). Individual strategies for
coping with stress at work: A review. Work & Stress, 7, 5-15.
Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542-
575.
Djupsjbacka, M., Johansson, H., Bergenheim, M., & Sjlander, P.
(1995). Influences on the -muscle-spindle system from
contralateral muscle afferents stimulated by KCl and lactic acid.
Neuroscience Research, 21, 301-309.
Djupsjbacka, M., Johansson, H., Bergenheim, M., & Wenngren, B. I.
(1995). Influences on the -muscle spindle system from muscle
afferents stimulated by increased intramuscular concentrations of
bradykinin and 5-HT. Neuroscience Research, 22, 325-333.
Doublet, S. (2000). The stress myth. Sydney: IPSILON Publishing.
Edwards, J. R., & Cooper, C. L. (1990). The person-environment fit
approach to stress: Recurring problems and some suggested
solutions. Journal of Organizational Behavior, 11, 293-307.
Elsass, P. M., & Veiga, J. F. (1997). Job control and job strain: A test of
three models. Journal of Occupational Health Psychology, 2, 195-211.
Eriksen, H. R., Ihlebk, C., & Ursin, H. (1999). A scoring system for
subjective health complaints. Scandinavian Journal of Public Health,
27, 63-72.
Eriksen, H. R., Svendsrd, R., Ursin, G., & Ursin, H. (1998). Prevalence
of subjective health complaints in the Nordic European countries in
1993. European Journal of Public Health, 8, 294-298.
Eriksen, H. R., & Ursin, H. (1999). Subjective health complaints: Is
coping more important than control? Work & Stress, 13, 238-252.
Eulberg, J. R., Weekley, J. A., & Bhagat, R. S. (1988). Models of stress in
organisational research: a metatheoretical perspective. Human
Relations, 41, 331-350.
Fletcher, B. C., & Jones, F. (1993). A refutation of Karaseks demand-
discretion model of occupational stress with a range of dependent
measures. Journal of Organizational Behavior, 14, 319-330.
Folkman, S. (1997). Positive psychological states and coping with severe
stress. Social Science & Medicine, 45, 1207-1221.
Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in middle-
aged community sample. Journal of Health and Social Behavior, 21,
219-239.
Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process:
Study of emotion and coping during three stages of a college
examination. Journal of Personality and Social Psychology, 54, 150-
170.

53
Folkman, S., & Lazarus, R. S. (1988). Coping as a mediator of emotion.
Journal of Personality and Social Psychology, 54, 466-475.
Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A., & Gruen,
R. (1986). The dynamics of a stressful encounter: Cognitive
appraisal, coping, and encounter outcomes. Journal of Personality
and Social Psychology, 50, 992-1003.
Folkman, S., Lazarus, R. S., Gruen, R., & DeLongis, A. (1986).
Appraisal, coping, health status, and psychological symptoms.
Journal of Personality and Social Psychology, 50, 572-579.
Fortes-Ferreira, L., Peir, J. M., Gonzlez-Morales, M. G., & Martin, I.
(2006). Work-related stress and well-being: The roles of direct
action coping and palliative coping. Scandinavian Journal of
Psychology, 47, 293-302.
French, J. R. P. (1973). Person role fit. Occupational Mental Health, 3,
15-20.
French, J. R. P., & Caplan, R. D. (1972). Organizational stress and
individual strain. In A. J. Marrow (Ed.), The failure of success (pp.
30-66). New York: Amacom.
French, J. R. P., Caplan, R. D., & Van Harrison, R. (1982). The
mechanisms of job stress and strain. Chichester: Wiley.
Frese, M., & Zapf, D. (1988). Methodological issues in the study of work
stress: Objective vs subjective measurement of work stress and the
question of longitudinal studies. In C. L. Cooper & R. Payne (Eds.),
Causes, coping, and consequences of stress at work (pp. 375-411). New
York: John Wiley & Sons Ltd.
Gerr, F., Marcus, M., Ensor, C., Kleinbaum, D., Cohen, S., Edwards, A.,
Gentry, E., Ortiz, D. J., & Monteilh, C. (2002). A prospective
study of computer users: I. Study design and incidence of
musculoskeletal symptoms and disorders. American Journal of
Industrial Medicine, 41, 221-235.
Glazer, S., & Beehr, T. A. (2005). Consistency of implications of three
role stressors across four countries. Journal of Organizational
Behavior, 26, 467-487.
Glorfeld, L. W. (1995). An improvement on Horns parallel analysis
methodology for selecting the correct number of factors to retain.
Educational and Psychological Measurement, 55, 377-393.
Grahn, B., Stigmar, K., & Ekdahl, C. (1999). Motivation for change in
patients with prolonged musculoskeletal disorders: A qualitative
two-year follow-up study. Physiotherapy Research International, 4,
170-189.

54
Greenglass, E. R. (2002). Work stress, coping and social support:
Implications for womens occupational well-being. In D. L. Nelson
& R. J. Burk (Eds.), Gender, work, stress and health (pp. 85-96).
Washington, DC.: American Psychological Association.
Greenglass, E. R., Burke, R. J., & Ondrack, M. (1990). A gender-role
perspective of coping and burnout. Applied Psychology: An
International Review, 39, 5-27.
Grier, K. S. (1982). A comparision of job stress in law enforcement and
teaching (Doctoral dissertation, University of South Florida, 1981).
Dissertation Abstracts International, 43, 870B.
Groenvold, M., Bjorner, J. B., Klee, M. C., & Kreiner, S. (1995). Test for
item bias in a quality of life questionnaire. Journal of Clinical
Epidemiology, 48, 805-816.
Grossi, G. (1999). Coping and emotional distress in a sample of Swedish
unemployed. Scandinavian Journal of Psychology, 40, 157-165.
Hackman, J. R., & Oldham, G. R. (1975). Development of the Job
Diagnostic Survey. Journal of Applied Psychology, 60, 159-170.
Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading:
Addison-Wesley.
Hagberg, M., Silverstein, B., Wells, R., Smith, M. J., Hendrick, H. W.,
Carayonn, P., & Perussen, M. (1995). Work related musculoskeletal
disorders (WMSDs). London: Taylor & Francis.
Harrison, R. V. (1978). Person-environment fit and job stress. In C. L.
Cooper & R. Payne (Eds.), Stress at work (pp. 175-205). New York:
Wiley.
Harrison, R. V. (1985). The person-environment fit model and the study
of job stress. In T. A. Beehr & R. S. Bhagat (Eds.), Human stress and
cognition in organizations (pp. 23-55). New York: Wiley.
Hartvigsen, J., Lings, S., Leboeuf-Yde, C., & Bakketeig, L. (2004).
Psychosocial factors at work in relation to low back bain and
consequences of low back pain; a systematic, critical review of
prospective cohort studies. Occupational and Environmental
Medicine, 61, 1-10.
Helmer, C., Barberger-Gateau, P., Letenneur, L., & Dartigues, J. F.
(1999). Subjective health and mortality in French elderly women
and men. Journals of Gerontology Series B: Psychological Sciences and
Social Sciences, 54, 84-92.
Helms, J. E., Henze, K. T., Sass, T. L., & Mifsud, V. A. (2006). Treating
Cronbachs alpha reliability coefficients as data in counselling
research. The Counseling Psychologist, 34, 630-660.

55
Hemingway, H., & Marmot, M. G. (1998). Psychosocial factors in the
primary and secondary prevention of coronary heart disease: A
systematic review. In S. Yusuf, J. Cairns, J. Cramm, E. Fallen & B.
Gersch (Eds.), Evidence based cardiology (pp. 269-285). London:
BMJ Publishing Group.
Hill, E. J., Martinson, V., Hawkins, A., & Ferris, M. (2003). Studying
working fathers: Comparing fathers and mothers work-family
conflict, fit, and adaptive strategies in a global high-tech company.
Fathering, 1, 239-261.
Hill, E. J., Miller, B. C., Weiner, S. P., & Colihan, J. (1998). Influences
of the virtual office on aspects of work and work/life balance.
Personnel Psychology, 51, 667-683.
Hobfoll, S. E., Dunahoo, C. L., Ben-Porath, Y., & Monnier, J. (1994).
Gender and coping: The dual-axis model of coping. American
Journal of Community Psychology, 22, 49-81.
Hodapp, V., Tanzer, N. K., Korunka, C., Maier, E. R., & Pestemer, I. A.
(2005). The German adaptation of the Job Stress Survey: A
multistudy validation in different occupational settings. In C. D.
Spielberger & I. Sarason (Eds.), Stress and emotion: Anxiety, anger,
and curiosity (Vol. 17, pp. 209-231). Washington, DC: Taylor and
Francis.
Holmstrm, S., Molander, B., & Barnekow-Bergqvist, M. (2004). Job
Stress Survey and self-reported health: Evaluating survey stability
through repeated measurements in metal-industry workplaces. In K.
B. Olsen & O. J. Teller (Eds.), Nordic Ergonomics Society 36th
Annual Conference NES2004 (pp. 239-242). Kolding, Denmark:
NES.
Holmstrm, S., Molander, B., Jansson, J., & Barnekow-Bergqvist, M.
(2008). Evaluation of a Swedish version of the Job Stress Survey.
Scandinavian Journal of Psychology, 49, 277-286.
Hooftman, W. E., van der Beek, A. J., Bongers, P. M., & van Mechelen,
W. (2005). Gender differences in self-reported physical and
psychosocial exposures in jobs with both female and male workers.
Journal of Occupational and Environmental Medicine, 47, 244-252.
Hoogendoorn, W. E., van Poppel, M. N., Bongers, P. M., Koes, B. W.,
& Bouter, L. M. (2000). Systematic review of psychosocial factors at
work and private life as risk factors for back pain. SPINE, 25, 2114-
2125.

56
Hurrell, J. J. J., Nelson, D. L., & Simmons, B. L. (1998). Measuring job
stressors and strains: Where we have been, where we are, and where
we need to go. Journal of Occupational Health Psychology, 3, 368-
389.
Hurst, T. E., & Hurst, M. H. (1997). Gender differences in mediation of
severe occupational stress among correctional officers. American
Journal of Criminal Justice, 22, 121-137.
Hgg, G. M. (1991). Static work loads and occupational myalgia - a new
explanation model. In P. A. Andersson, D. J. Hobart & J. V. Danoff
(Eds.), Electromyographical kinesiology (pp. 141-144). Amsterdam:
Elsevier Science Publishers.
Hgg, G. M., & strm, A. (1997). Load pattern and pressure pain
threshold in the upper trapezius muscle and psychosocial factors in
medical secretaries with and without shoulder/neck disorders.
International Archives of Occupational and Environmental Health, 69,
423-432.
Ihlebk, C., Eriksen, H. R., & Ursin, H. (2002). Prevalence of subjective
health complaints (SHC) in Norway. Scandinavian Journal of Public
Health, 30, 20-29.
Jackson, S. E., & Schuler, R. S. (1985). A meta-analysis and conceptual
critique of research on role ambiguity and role conflict in work
settings. Organizational Behavior and Human Decision Processes, 36,
16-78.
Jodoin, M. G., & Gierl, M. J. (2001). Evaluating type I error and power
rates using an effect size measure with the logistic regression
procedure for DIF detection. Applied Measurement in Education, 14,
329-349.
Johansson, H., & Sojka, P. (1991). Pathophysiological mechanisms
involved in genesis and spread of muscular tension in occupational
muscule pain and in chronic musculoskeletal pain syndromes: a
hypothesis. Medical Hypotheses, 35, 196-203.
Johansson, J. . (1994). Psychosocial factors at work and their relation to
musculoskeletal symptoms. Doctoral dissertation: Gteborg
University, Sweden.
Johnson, J. V. (1989). Control, collectivity and the psychosocial work
environment. In S. L. Sauter, J. J. Hurrell & C. L. Cooper (Eds.),
Job control and worker health. Chichester: Wiley.
Johnson, J. V., & Hall, E. M. (1988). Job strain, work place social
support, and cardiovascular disease: A cross-sectional study of a
random sample of the Swedish working population. American
Journal of Public Health, 78, 1336-1342.

57
Johnson, J. V., Hall, E. M., & Theorell, T. (1989). Combined effects of
job strain and social isolation on cardiovascular disease morbidity
and mortality in a random sample of the Swedish male working
population. Scandinavian Journal of Work, Environment & Health,
15, 271-279.
Kaiser, H. F. (1960). The application of electronic computers to factor
analysis. Educational and Psychological Measurement, 20, 141-151.
Karasek, R. A. (1979). Job demands, job decision latitude, and mental
strain: Implications for job redesign. Administrative Science
Quarterly, 24, 285-307.
Karasek, R. A. (1985). Job content questionnaire and users guide. Los
Angeles: University of Southern California Department of Industrial
and Systems Engineering.
Karasek, R. A. (1998). Demand/Control model: A social, emotional and
physiological approach to stress risk and active behavior
development. In J. M. Stellman (Ed.), Encyclopaedia of occupational
health and safety (pp. 34.36-34.14). Geneva: ILO.
Karasek, R. A., Brisson, C., Kawakami, N., Houtman, I., & Bongers, P.
M. (1998). The Job Content Questionnaire (JCQ): An instrument
for internationally comparative assessments of psychosocial job
characteristics. Journal of Occupational Health Psychology, 3, 322-
355.
Karasek, R. A., & Theorell, T. (1990). Healthy work: Stress, productivity,
and the reconstruction of working life. New York: Basic Books.
Kasl, S. V. (1978). Epidemiological contributions to the study of work
stress. In C. L. Cooper & R. L. Payne (Eds.), Stress at work (pp. 3-
38). New York: Wiley.
Kasl, S. V. (1987a). Methodologies in stress and health: Past difficulties,
present dilemmas, future directions. In S. V. Kasl & C. L. Cooper
(Eds.), Research methods in stress and health psychology (pp. 307-318).
New York: John Wiley & Sons.
Kasl, S. V. (1987b). The researcher's perspective. In J. C. Quick, R. S.
Bhagat, J. E. Dalton & J. D. Quick (Eds.), Work stress: Health care
systems in the workplace (pp. 16-20). New York: Praeger.
Kasl, S. V. (1998). Measuring job stressors and studying the health impact
of the work environment: An epideminologic commentary. Journal
of Occupational Health Psychology, 3, 390-401.
Kawakami, N., Haratani, T., & Araki, S. (1992). Effects of perceived job
stress on depressive symptoms in blue-collar workers of an electrical
factory in Japan. Scandinavian Journal of Work, Health &
Environment 18, 195-200.

58
Kilbom, A., Armstrong, T., Buckle, P., Fine, L., Hagberg, M., Haring-
Sweeney, M., Martin, B., Punnett, L., Silverstein, B., Sjgaard, G.,
Theorell, T., & Viikari-Juntura, E. (1996). Musculoskeletal
disorders: work-related risk factors and prevention. International
Journal of Occupational and Environmental Health, 2, 239-246.
Kjellberg, A., & Wadman, C. (2007). The role of the affective stress
response as a mediator of the effect of psychosocial risk factors on
musculoskeletal complaints - Part 1: Assembly workers. International
Journal of Industrial Ergonomics, 37, 367-374.
Knutson, G. A. (2000). The role of the -motor system in increasing
muscle tone and muscle pain syndromes: A review of the
Johansson/Sojka hypothesis. Journal of Manipulative and
Physiological Therapeutics, 23, 564-572.
Korunka, C., Weiss, A., Huemer, K.-H., & Karetta, B. (1995). The
effects of new technologies on job satisfaction and psychosomatic
complaints. Applied Psychology: An International Review, 44, 123-
142.
Krajewski, H. T., & Goffin, R. D. (2005). Predicting occupational coping
responses: The interactive effect of gender and work stressor context.
Journal of Occupational Health Psychology, 10, 44-53.
Krause, N., Lynch, J., Kaplan, G. A., Cohen, R. D., Goldberg, D. E., &
Salonen, J. T. (1997). Predictors of disability retirement.
Scandinavian Journal of Work, Environment & Health, 23, 403-413.
Krause, N., Ragland, D. R., Greiner, B. A., Syme, S. L., & Fisher, J. M.
(1997). Psychosocial job factors associated with back and neck pain
in public transit operators. Scandinavian Journal of Work,
Environment & Health, 23, 179 186.
Kristensen, T. S. (1995). The demand-control-support model:
Methodological challenges for future research. Stress Medicine, 11,
17-26.
Kristensen, T. S. (1996). Job stress and cardiovascular disease: A
theoretical critical review. Journal of Occupational Health Psychology,
1, 246-260.
Kristensen, T. S., Bjorner, J. B., Christensen, K. B., & Borg, V. (2004).
The distinction between work pace and working hours in the
measurement of quantitative demands at work. Work & Stress, 18,
305-322.
Kristjansson, E., Aylesworth, R., McDowell, I., & Zumbo, B. D. (2005).
A comparison of four methods for detecting differential item
functioning in ordered response items. Educational and Psychological
Measurement, 65, 935-953.

59
Kushnir, T., & Melamed, S. (1991). Work-load, perceived control and
psychological distress in Type A/B industrial workers. Journal of
Organizational Behavior, 12, 155-168.
Larsson, B., Sgaard, K., & Rosendal, L. (2007). Work related neck-
shoulder pain: A review on magnitude, risk factors, biochemical
characteristics, clinical picture and preventive interventions. Best
Practice & Research Clinical Rheumatology, 21, 447-463.
Lazarus, R. S. (1966). Psychological stress and the coping process. New York:
McGraw-Hill.
Lazarus, R. S. (1977). Psychological stress and coping in adaptation and
illness. In Z. J. Lipowski, D. R. Lipsitt & P. C. Whybrow (Eds.),
Psychosomatic medicine: Current trends and clinical applications (pp.
14-26). New York: Oxford University Press.
Lazarus, R. S. (1991). Psychological stress in the workplace. Journal of
Social Behavior and Personality, 6, 1-13.
Lazarus, R. S. (1993). From psychological stress to the emotions: A
history of changing outlooks. Annual Review of Psychology, 44, 1-21.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New
York: Springer.
Lazarus, R. S., & Folkman, S. (1987). Transactioal theory and research on
emotions and coping. European Journal of Personality, 1, 141-170.
Leino-Arjas, P., Hnninen, K., & Puska, P. (1998). Socioeconomic
variation in back and joint pain in Finland. European Journal of
Epidemiology, 14, 7987.
Leino, P., & Hanninen, V. (1995). Psychosocial factors at work in
relation to back and limb disorders. Scandinavian Journal of Work,
Environment & Health, 21, 134-142.
Leiter, M. P. (1991). Coping patterns as preditors of burnout: The
function of control and escapist coping patterns. Journal of
Organizational Behavior, 12, 123-144.
Levenstein, S., Prantera, C., Varvo, V., Scribano, M. L., Berto, E., Luzi,
C., & Andreoli, A. (1993). Development of the perceived stress
questionnaire: A new tool for psychosomatic research. Journal of
Psychosomatic Research, 37, 19-32.
Linton, S. J. (2001). Occupational psychological factors increase the risk
for back pain: A systematic review. Journal of Occupational
Rehabilitation, 11, 53-66.
Linton, S. J., Hellsing, A. L., & Halldn, K. (1998). A population-based
study of spinal pain among 3545-year-old individuals. Prevalence,
sick leave, and health care use. SPINE, 23, 14571463.

60
Liu, C., Spector, P. E., & Shi, L. (2007). Cross-national job stress: a
quantitative and qualitative study. Journal of Organizational
Behavior, 28, 209-239.
Locke, E. A. (1969). What is job satisfaction? Organizational Behavior and
Human Performance, 4, 309-336.
Locke, E. A. (1976). The nature and causes of job satisfaction. In M.
Dunnette (Ed.), Handbook of industrial and organizational psychology
(pp. 1297-1350). Chicago: Rand McNally.
Loewenthal, K. M. (1996). An introduction to psychological tests and scales.
London: UCL Press.
Long, B. C., & Kahn, S. E. (1993). A theoretical integration of women,
work and coping. In B. C. Long & S. E. Kahn (Eds.), Women, work
and coping (pp. 296-311). Montreal: McGill-Queens University
Press.
Love, K. G., & Beehr, T. A. (1981). Social stressors on the job:
Recommendations for a broadened perspective. Group &
Organization Studies, 6, 190-200.
Lundberg, U. (2006). Stress, subjective and objective health. International
Journal of Social Welfare, 15, 41-48.
Lundberg, U., Forsman, M., Zachau, G., Eklf, M., Palmerud, G., Melin,
B., & Kadefors, R. (2002). Effects of experimentally induced mental
and physical stress on motor unit recruitment in the trapezius
muscle. Work & Stress, 16, 166-178.
Lundberg, U., Kadefors, R., Melin, B., Palmerud, G., Hassmn, P.,
Engstrm, M., & Elfsberg Dohns, I. (1994). Psychophysiological
stress and EMG activity of the Trapezius muscle. International
Journal of Behavioral Medicine, 1, 354-370.
Major, V. S., Klein, K. S., & Ehrhart, M. G. (2002). Work time,work
interference with family and psychological distress. Journal of
Applied Psychology, 87, 427-436.
Makin, P. J., Cooper, C. L., & Cox, C. J. (1996). Organizations and the
psychological contract. Leicester: British Psychological Society.
Makowska, Z. (1995). Psychosocial characteristics of work and family as
determinants of stress and well-being of women: A preliminary
study. International Journal of Occupational Medicine and
Environmental Health, 8, 215-222.
Marhold, C., Linton, S. J., & Melin, L. (2002). Identification of obstacles
for chronic pain patients to return to work: Evaluation of a
questionnaire. Journal of Occupational Rehabilitation, 12, 65-75.

61
Marmot, M. G., & Madge, N. (1987). An epidemiological perspective on
stress and health. In S. V. Kasl & C. L. Cooper (Eds.), Stress and
health: Issues in research methodology (pp. 3-26). Chichester: Wiley.
Marmot, M. G., Siegrist, J., Theorell, T., & Feeney, A. (1999). Health
and the psychosocial environment at work. In M. G. Marmot & R.
Wilkinson (Eds.), Social determinants of health. Oxford: Oxford
University Press.
Mason, J. W. (1975a). A historical view of the stress field. Part I. Journal
of Human Stress, 1, 6-12.
Mason, J. W. (1975b). A historical view of the stress field. Part II. Journal
of Human Stress, 1, 22-36.
Mauno, S., & Kinnunen, U. (1999). The effects of job stressors on
marital satisfaction in Finnish dual-earner couples. Journal of
Organizational Behavior, 20, 879-895.
Melamed, S., Benavi, I., Luz, J., & Green, M. S. (1995). Repetitive work,
work underload and coronary heart-disease risk-factors among blue-
collar workers - the cordis study. Journal of Psychosomatic Research,
39, 19-29.
Melin, B., & Wigaeus Trnqvist, E. (2005). Can the psychosocial
environment cause neck and shoulder pain? In R. . Gustafsson &
I. Lundberg (Eds.), Work life and health in Sweden 2004 (pp. 131-
157). Stockholm: National Institute for Working Life & Swedish
Work Environment Authority.
Miilunplao, S., Vuori, I., Oja, P., Pasanen, M., & Urponen, H. (1997).
Self-rated health as a health measure: The predictive value of self-
reported health status on the use of physician services and mortality
in the working-age population. Journal of Clinical Epidemiology, 50,
517-528.
Molander, B., Holmstrm, S., & Jansson, J. (2007). Svenska normer fr
Job Stess Survey. Metalindustri och sjukvrd [Swedish norms for the Job
Stress Survey. Metal industry and medical service]. Ume University:
Department of Psychology.
Motowidlo, S. J., Packard, J. S., & Manning, M. R. (1986). Occupational
stress - Its causes and consequences for job-performance. Journal of
Applied Psychology, 71, 618-629.
Moyle, P. (1998). Longitudinal influences of managerial support on
employee well-being. Work and Stress, 12, 29-49.
Murphy, L. R. (1996). Stress management in work settings: A critical
review of the health effects. American Journal of Health Promotion,
11, 112-135.

62
Murphy, L. R., & Hurrell, J. J. (1987). Stress measurement and
management in organizations: Development and current status. In
A. W. Riley & S. J. Zaccharo (Eds.), Occupational stress and
organizational effectiveness (pp. 29-51). New York: Praeger.
Murphy, L. R., Hurrell, J. J., Sauter, S. L., & Keita, G. P. (1995). Job
stress interventions. Washington, DC: American Psychological
Association.
Narayanan, L., Menon, S., & Spector, P. E. (1999). Stress in the
workplace: A comparison of gender and occupations. Journal of
Organizational Behavior, 20, 63-73.
Naylor, J. C., Pritchard, R. D., & Ilgen, D. R. (1980). A theory of behavior
in organizations. New York: Academic Press.
Nilsson, L-G., Adolfsson, R., Backman, L., de Frias, C. M., Molander, B.,
& Nyberg, L. (2004). Betula: A prospective cohort study on
memory, health and aging. Aging, Neuropsychology, and Cognition,
11, 134-148.
Nilsson, L-G., Backman, L., Erngrund, K., Nyberg, L., Adolfsson, R.,
Bucht, G., Karlsson, S., Widing, M., & Winblad, B. (1997). The
Betula prospective cohort study: Memory, health and aging. Aging,
Neuropsychology, and Cognition, 4, 1-32.
Nolan-Hoeksema, S. (1990). Sex differences in depression. Stanford, CA:
Stanford University Press.
Nordander, C., Ohlsson, K., Balogh, I., Rylander, L., Plsson, B., &
Skerfving, S. (1999). Fish processing work: the impact of two sex
dependent exposure profiles on musculoskeletal health. Occupational
and Environmental Medicine, 56, 256 264.
Nunnally, J., & Bernstein, I. (1994). Psychometric theory. New York:
McGraw-Hill.
Nystedt, L., Sjberg, A., & Hgglund, G. (1999). Discriminant validation
of measures of organizational commitment, job involvement, and
job satisfaction among Swedish army officers. Scandinavian Journal
of Psychology, 40, 49-55.
O'Driscoll, M. P. (1996). The interface between job and off-job roles:
Enhancement and conflict. International Review of Industrial and
Organizational Psychology, 11, 279-306.
O'Driscoll, M. P., & Beehr, T. A. (1994). Supervisor behaviors, role
stressors and uncertainty as predictors of personal outcomes for
subordinates. Journal of Organizational Behavior, 15, 141-155.

63
ORoark, A. M. (1995). Occupational stress and informed interventions.
In C. D. Spielberger, I. G. Sarason, J. M. T. Brebner, E. Greenglass,
P. Laungani & A. M. ORoark (Eds.), Stress and emotion: Anxiety,
anger, and curiosity (Vol. 15, pp. 121-135). Washington, DC:
Taylor & Francis.
Osipow, S., & Spokane, A. (1987). Manual for Occupational Stress
Inventory: Research version. Odessa FL: Psychological Assessment
Resources.
Osterlind, S. J. (1983). Test item bias. Oxford: Oxford University Press.
Parasuraman, S., & Greenhaus, J. H. (2002). Toward reducing some
critical gaps in workfamily research. Human Resource Management
Review, 12, 299-312.
Parkes, K. R. (1990). Coping, negative affectivity, and the work
environment: Additive and interactive predictors of mental health.
Journal of Applied Psychology, 75, 399-409.
Passatore, M., & Roatta, S. (2003). Sympathetic nervous system:
Interaction with muscle function and involvement in motor control.
In H. Johansson, U. Windhorst, M. Djupsjbacka & M. Passatore
(Eds.), Chronic work-related myalgia. Neuromuscular mechanisms
behind work-related chronic muscle pain syndromes (pp. 243-264).
Gvle, Sweden: Gvle University Press.
Pedersen, J., Sjlander, P., Wenngren, B. I., & Johansson, H. (1997).
Increased intramuscular concentration of bradykinin increases the
static fusimotor drive to muscle spindles in neck muscles of the cat.
Pain, 70, 83-91.
Peter, R., Alfredsson, L., Hammar, N., Siegrist, J., Theorell, T., &
Westerholm, P. (1998). High effort, low reward and cardiovascular
risk factors in employed Swedish men and women: Baseline results
from the WOLF Study. Journal of Epidemiology and Community
Health, 52, 540-547.
Peter, R., Geiler, H., & Siegrist, J. (1998). Associations of effort-reward
imbalance at work and reported symptoms in different groups of
male and female public transport workers. Stress Medicine, 14, 175-
182.
Peter, R., & Siegrist, J. (1997). Chronic work stress, sickness absence and
hypertension in middle managers: General or specific sociological
explanations? Social Science & Medicine, 45, 1111-1120.

64
Petersen, M. A., Groenvold, M., Bjorner, J. B., Aaronson, N., Conroy,
T., Cull, A., Fayers, P., Hjermstad, M., Sprangers, M., & Sullivan,
M. (2003). Use of differential item functioning analysis to assess the
equivalence of translations of a questionnaire. Quality of Life
Research, 12, 373-385.
Peterson, M. F., & Smith, P. B. (1997). Does national culture or ambient
temperature explain corss-cultural differences in role stress? No swea!
Academy of Management Journal, 40, 930-947.
Punnett, L., & Gold, J. E. (2003). Work-related upper extremity
disorders: Epidemiologic findings and unresolved questions. In H.
Johansson, U. Windhorst, M. Djupsjbacka & M. Passatore (Eds.),
Chronic Work-Related Myalgia (pp. 47-56). Gvle: Gvle University
Press.
Punnett, L., & Herbert, R. (2000). Work-related musculoskeletal
disorders: Is there a gender differential, and if so, what does it mean?
In M. Goldman & M. Hatch (Eds.), Women and health (pp. 474-
491). San Diego: Academic Press.
Quick, J. C., & Quick, J. D. (1984). Organizational stress and preventive
management. New York: McGraw-Hill.
Quick, J. C., Quick, J. D., Nelson, D. L., & Hurrell, J. J. (1997).
Preventive stress management in organizations. Washington, DC:
American Psychological Association.
Rick, J., Briner, R. B., Daniels, K., Perryman, S., & Guppy, A. (2001). A
critical review of psychosocial hazard measures. Sudbury: HSE Books.
Rick, J., & Guppy, A. (1994). Coping strategies and mental health in
white collar public sector employees. European Work and
Organizational Psychologist, 4, 121-137.
Rydstedt, L. W., Devereux, J., & Sverke, M. (2007). Comparing and
combining the demand-control-support model and the effort reward
imbalance model to predict long-term mental strain. European
Journal of Work and Organizational Psychology, 16, 261-278.
Sandsj, L., Melin, B., Rissn, D., Dohns, I., & Lundberg, U. (2000).
Trapezius muscle activity, neck and shoulder pain, and subjective
experiences during monotonous work in women. European Journal
of Applied Physiology, 83, 235-238.
Schnall, P. L., Landisbergis, P. A., & Baker, D. (1994). Job strain and
cardiovascular disease. Annual Review of Public Health, 15, 381-411.
Schuler, R. S. (1980). Definition and conceptualization of stress in
organizations. Organizational Behavior and Human Performance, 25,
184-215.

65
Scott, D. B. (2001). The costs and benefits of womens family ties in
occupational context: Women in corporate-government affairs
management. Community, Work and Family, 4, 5-27.
Selye, H. (1936). A syndrome produced by diverse noxious agents.
Nature, 138, 32.
Selye, H. (1946). The general adaptation syndrome and diseases of
adaptation. Journal of Clinical Endocrinology, 6, 217-230.
Selye, H. (1967). In vivo. The case for supramolecular biology. New York:
Liveright.
Selye, H. (1974). Stress without distress. Philadelphia: J.B Lippincott Co.
Selye, H. (1983). The stress concept: Past, present and future. In C. L.
Cooper (Ed.), Stress research: Issues for the eighties (pp. 1-20).
Chichester: John Wiley and Sons.
Seyle, H. (1975). Confusion and controversy in the stress field. Journal of
Human stress, 1, 37-44.
Seyle, H. (1976). The stress of life. New York: McGraw-Hill.
Sharit, J., & Salvendy, G. (1982). Occupational stress: Review and
reappraisal. Human Factors, 24, 129-162.
Siegrist, J. (1996). Adverse health effects of high-effort/low-reward
conditions. Journal of Occupational Health Psychology, 1, 27-41.
Siegrist, J. (1998). Adverse health effects of effort-reward imbalance at
work: Theory, empirical support, and implications for prevention.
In C. L. Cooper (Ed.), Theories of organizational stress (pp. 190-
204). Oxford: Oxford University Press.
Siegrist, J., & Marmot, M. G. (2004). Health inequalities and the
psychosocial environment - two scientific challenges. Social Science
& Medicine, 58, 1463-1499.
Siivola, S. M., Levoska, S., Latvala, K., Hoskio, E., Vanharanta, H., &
Keinnen-Kiukaanniemi, S. (2004). Predictive factors for neck and
shoulder pain: A longitudinal study in young adults. SPINE, 29,
16621669.
Sjgaard, G., Lundberg, U., & Kadefors, R. (2000). The role of muscle
activity and mental load in the development of pain and
degenerative processes at the muscle cell level during computer
work. European Journal of Applied Physiology, 83, 99-105.
Smri, J., Arason, E., Hafsteinsson, H., & Ingimarsson, S. (1997).
Unemployement, coping and psychological distress. Scandinavian
Journal of Psychology, 38, 151-156.
Smith, C., & Lazarus, R. S. (1990). Emotion and adaption. In L. A.
Pervin (Ed.), Handbook of personality: Theory and research (pp. 609-
637). New York: Guilford Press.

66
Smith, L. L. (2002). On the usefulness of item bias analysis to personality
psychology. Personality and Social Psychology Bulletin, 28, 754-763.
Sparks, K., Cooper, C. L., Fried, Y., & Shirom, A. (1997). The effects of
work hours on health: A meta-analytic review. Journal of
Occupational and Organizational Psychology, 70, 389-400.
Spector, P. E., Cooper, C. L., Poelmans, S., Allen, T. D., ODriscoll, M.,
Sanchez, J. I., Siu, O., Dewe, P. J., Hart, P., Lu, L., De-Moreas, L.
F. R., Ostroganay, G. M., Sparks, K., Wong, P. T. P., & Yu, S.
(2004). A cross-national comparative study of work-family stressors,
working hours, and well-being: China and Latin America vs. the
Anglo world. Personnel Psychology, 57, 119-142.
Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory
(Form Y). Palo Alto, CA.: Consulting Psychologists Press.
Spielberger, C. D. (1988). Manual for the State-Trait Anger Expression
Inventory (STAXI). Oddessa, FL: Psychological Assessment
Resources.
Spielberger, C. D. (1991). Preliminary test manual for the Job Stress Survey
(JSS). Odessa, FL: Psychological Assessment Resources.
Spielberger, C. D., Grier, K. S., & Pate, J. M. (1980). The Florida police
stress survey. Florida Fraternal Order of Police Journal, 66-67.
Spielberger, C. D., & Reheiser, E. C. (1994a). Job stress in university,
coporate, and military personal. International Journal of Stress
Management, 1, 19-31.
Spielberger, C. D., & Reheiser, E. C. (1994b). The Job Stress Survey -
Measuring gender differences in occupational stress. Journal of Social
Behavior and Personality, 9, 199-218.
Spielberger, C. D., Reheiser, E. C., Reheiser, J. E., & Vagg, P. R. (2000).
Measuring stress in the workplace: The Job Stress Survey. In D. T.
Kenny, J. G. Carlson, F. J. McGuigan & J. L. Sheppard (Eds.),
Stress and Health: Research and Clinical Applications (Vol. 1, pp. 397-
409). Sydney: Harwood/Gordon & Breach Science International.
Spielberger, C. D., & Vagg, P. R. (1999). Job Stress Survey. Professional
Manual. Odessa, FL: Psychological Assessment Resources.
Spielberger, C. D., Westberry, L. G., Grier, K. S., & Greenfield, G.
(1981). The Police Stress Survey: Sources of stress in law enforcement.
(Human Resources Institute Monograph Series Three, No. 6).
Tampa, FL: University of South Florida, College of Social and
Behavioral Sciences.

67
Stansfeld, S. A., Bosma, H., Hemingway, H., & Marmot, M. G. (1998).
Psychosocial work characteristics and social support as predictiors of
SF-36 health functioning: The Whithall II study. Psychosomatic
Medicine, 60, 247-255.
Statistics Sweden. (2006). Arbetsorsakade besvr 2006 [Work-related
disorders 2006. Summary in English]: Sveriges officiella statistik.
Statistiska meddelanden AM 43 SM 0601. Stockholm: SCB.
Stevens, J. (2002). Applied multivariate statistics for the social sciences.
Mahwah, NJ: Lawrence Erlbaum.
Swaminathan, H., & Rogers, H. J. (1990). Detecting differential item
functioning using logistic regression procedures. Journal of
Educational Measurement, 27, 361-370.
Tamres, L. K., Janicki, D., & Helgeson, V. S. (2002). Sex differences in
coping behavior: A meta-analytic review and an examination of
relative coping. Journal of Personality and Social Psychology Review, 6,
2-30.
Taris, T. W., & Kompier, M. A. J. (2003). Challenges in longitudinal
designs in occupational health psychology. Scandinavian Journal of
Work, Environment & Health, 29, 1-4.
Tinsley, H. E. A., & Tinsley, D. J. (1987). Use of factor analysis in
counseling psychology research. Journal of Counseling Psychology, 34,
414-424.
Treaster, D., E., & Burr, D. (2004). Gender differences in prevalence of
upper extremity musculoskeletal disorders. Ergonomics, 47, 495-526.
Triosi, A. (2001). Gender differences in vulnerability to social stress: A
Darwinian perspective. Physiology & Behavior, 73, 443-449.
Tsutsumi, A., & Kawakami, N. (2004). A review of empirical studies on
the model of effort-reward imbalance at work: Reducing
occupational stress by implementing a new theory. Social Science &
Medicine, 59, 2335-2359.
Turnage, J. J., & Spielberger, C. D. (1991). Job stress in managers,
professionals, and clerical workers. Work and Stress, 5, 165-176.
Vagg, P. R., & Spielberger, C. D. (1998). Occupational stress: Measuring
job pressure and organizational support in the workplace. Journal of
Occupational Health Psychology, 3, 294-305.
Vagg, P. R., & Spielberger, C. D. (1999). The Job Stress Survey:
Assessing perceived severity and frequency of occurrence of generic
sources of stress in the workplace. Journal of Occupational Health
Psychology, 4, 288-292.

68
Vagg, P. R., Spielberger, C. D., & Wasala, C. F. (2002). Effects of
organizational level and gender on stress in the workplace.
International Journal of Stress Management, 9, 243-261.
van der Doef, M., & Maes, S. (1998). The job demand-control(-support)
model and physical health outcomes: A review of the strain and
buffer hypotheses. Psychology and Health, 13, 909-936.
van der Doef, M., & Maes, S. (1999). The job demand-control(-support)
model and psychological well-being: A review of 20 years of
empirical research. Work & Stress, 13, 87-114.
van der Windt, D. A. W. M., Thomas, E., Pope, D. P., de Winter, A. F.,
Macfarlane, G. J., Bouter, L. M., & Silman, A. J. (2000).
Occupational risk factors for shoulder pain: a systematic review.
Occupational and Environmental Medicine, 57, 433-442.
van Dien, J. H., Visser, B., & Hermans, V. (2003). The contribution of
task-related biomechanical constraints to the development of work-
related myalgia. In H. Johansson, U. Windhorst, M. Djupsjbacka
& M. Passatore (Eds.), Chronic Work-Related Myalgia (pp. 83-93).
Gvle: Gvle University Press.
van Kempen, E. E. M. M., Kruize, H., Boshuizen, H. C., Ameling, C. B.,
Staatsen, B. A. M., & de Hollander, A. E. M. (2002). The
association between noise exposure and blood pressure and ischemic
heart disease: A meta-analysis. Environmental Health Perspectives,
110, 307-317.
Van Vegchel, N., de Jonge, J., Bosma, H., & Schaufeli, W. (2005).
Reviewing the effort-reward imbalance model: Drawing up the
balance of 45 empirical studies. Social Science & Medicine, 60, 1117-
1131.
Warr, P. (1994). A conceptual framework for the study of work and
mental health. Work & Stress, 8, 84-97.
Weaver, K. E., Antoni, M. H., Lechner, S. C., Durn, R. E. F., Penedo,
F., Fernandez, M. I., Ironson, G., & Schneiderman, N. (2004).
Perceived stress mediates the effects of coping on the quality of life
of hiv-positive women on highly active antiretroviral therapy. AIDS
and Behavior, 8, 175-183.
Wenngren, B. I., Pedersen, J., Sjlander, P., Bergenheim, M., &
Johansson, H. (1998). Bradykinin and muscle stretch alter
contralateral cat neck muscle spindle output. Neuroscience Research,
32, 199-129.
Verhoeven, C., Maes, S., Kraaij, V., & Joekes, K. (2003). The job
demand-control-social support model and wellness/health outcomes:
A european study. Psychology and Health, 18, 421-440.

69
Westman, M., & Eden, D. (1992). Excessive role demand and subsequent
performance. Journal of Organizational Behavior, 13, 519-529.
Williams, S., & Cooper, C. L. (1998). Measuring occupational stress:
Development of the pressure management indicator. Journal of
Occupational Health Psychology, 3, 306-321.
Wong, P. T. P. (1993). Effective management of life stress: The resource-
congruence model. Stress Medicine, 9, 51-60.
Woods, V. (2005). Work-related musculoskeletal health and social
support. Occupational Medicine, 55, 17-189.
Wright, L. A., & Smye, M. D. (1996). Corporate abuse: How lean and
mean robs people and profits. New York: Macmillan.
Zautra, A. J. (2003). Emotions, stress, and health. New York: Oxford
University Press.
Zetterberg, C., Forsberg, A., Hansson, E., Johansson, H., Nielsen, P.,
Danielsson, B., Inge, G., & Olsson, B. (1997). Neck and upper
extremity problems in car assembly workers. A comparison of
subjective complaints, work satisfaction, physical examination and
gender. International Journal of Industrial Ergonomics, 19, 277-289.
Zumbo, B. D. (1999). A handbook on the theory and methods of differential
item functioning (DIF): Logistic regression modeling as a unitary
framework for binary and likert-type (ordinal) item scores. Ottawa:
Directorate of Human Resources Research and Evaluation,
Department of National Defense, Canada.

70